(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Experimental and theoretical investigation of the reactivity of partially fluorinated radicals"

EXPERIMENTAL AND THEORETICAL INVESTIGATION OF THE REACTIVITY 
OF PARTIALLY FLUORINATED RADICALS 






By 

MICHAEL DAVID BARTBERGER 



A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL 

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT 

OF THE REQUIREMENTS FOR THE DEGREE OF 

DOCTOR OF PHILOSOPHY 



UNIVERSITY OF FLORIDA 
1998 



ACKNOWLEDGEMENTS 

Among the great number of individuals with whom I have interacted throughout 
the course of my education at the University of Florida and elsewhere, I wish to express 
my sincere appreciation to the special few that have motivated, challenged, and inspired 
me. 

I extend my deepest gratitude to Prof. William R. Dolbier, Jr., an outstanding 
scientist and truly exceptional educator, for his excellent guidance, support, and 
friendship throughout the course of my graduate career. My appreciation for the 
knowledge he has shared with me, as well as his patience and level of understanding, 
particularly during periods of difficulty and stress, can not be overstated. 

I wish to thank the two finest classroom instructors I have ever had-my first 
college level chemistry teacher, Dr. Jeanette Madea, for her profound influence in my 
decision to pursue a career in the chemical sciences, and Prof. Seth Elsheimer, for my 
initial exposure to the fascinating area of organofluorine chemistry in 1990 and his 
friendship thereafter. 

I am indebted to Dr. Max Muir for introducing me to computational chemistry. 
The experience I have gained in the use of molecular orbital methods as a tool for the 
understanding of chemical reactivity is due entirely to him. Special thanks go to Prof. 
Benjamin Horenstein for his helpful discussions and generosity with regard to 
computational resources. 

I thank my colleagues, past and present, in both the Dolbier group and the 
Department of Chemistry as a whole. A few bear special mention-Dr. Keith Palmer, for 
his friendship and advice during my first year in the group; Dr. Xiao Xin Rong and He-Qi 



Pan, for their camaraderie and early assistance with radical kinetics; Dr. Conrad 
Burkholder, for numerous stimulating discussions; and Dr. Henryk Koroniak, Michelle 
Fletcher, Lian Luo, Feng Tian, and Kevin Ley for their friendship (and tolerance!) 
throughout the course of my stay in the department. 

I wish to thank my graduate committee, particularly the "organic" portion thereof- 
Profs. Merle Battiste and Kirk Schanze, for their advice and encouragement. Also, 
special thanks go to Prof. R. J. Bartlett for taking seriously my interest in theoretical 
methodology and the invitations to participate in his workshops on Applied Molecular 
Orbital Theory. 

I am especially grateful to my very best friend, Cynthia Dawn Zook, for her 
unrelenting moral support and encouragement over the last several years. Finally, I 
wish to acknowledge my parents, George Charles and Beverly Jean, for instilling in me 
the work ethic which has likely had as much to do with the successful completion of this 
work as any of the chemistry I ever learned. 



in 



TABLE OF CONTENTS 

ACKNOWLEDGEMENTS N 

ABSTRACT vi 

CHAPTER 

1 AN OVERVIEW OF ORGANIC FREE RADICAL REACTIONS 1 

Introduction 1 

Radical Chain Processes 2 

Hydrogen Atom Abstraction Reactions 4 

Intermolecular Radical Addition Reactions 9 

Intramolecular Addition Reactions: Radical Cyclizations 13 

Methods for Determination of Organic Radical Kinetics 22 

Conclusion 26 

2 THE FLUORINE SUBSTITUENT IN ORGANIC SYSTEMS 28 

Introduction 28 

Structure, Bonding, and Reactivity in Saturated Systems 29 

Structure, Bonding, and Reactivity in Unsaturated Systems 31 

Fluorine Non-Bonded Interactions in Reactive Intermediates 33 

Fluorine Steric Effects 35 

The Fluorine Substituent in Free Radicals 37 

Organofluorine Radical Reactivity 40 

Conclusion 48 

3 THE REACTIVITY OF PARTIALLY FLUORINATED RADICALS IN 
INTERMOLECULAR ADDITION AND HYDROGEN ABSTRACTION 
REACTIONS 49 

Introduction 49 

Precursor Syntheses and Competitive Kinetic Studies 50 

Discussion 61 

Conclusion 76 

4 THE REACTIVITY OF PARTIALLY FLUORINATED RADICALS IN 
INTRAMOLECULAR CYCLIZATION REACTIONS 77 

Introduction 77 

Precursor Syntheses and Competitive Kinetic Studies 78 

Discussion 89 

Conclusion 100 



IV 



5 EXPERIMENTAL 102 

General Methods- Experimental 102 

General Methods- Theoretical 103 

Synthetic Procedures 103 

Competitive Kinetic Procedures 129 

APPENDIX A: SELECTED 19 F NMR SPECTRA 136 

APPENDIX B: B3LYP/6-31G(d) TOTAL AND ZERO-POINT ENERGIES FOR 

DATA IN TABLES 3-3 AND 3-4 172 

REFERENCES 175 

BIOGRAPHICAL SKETCH 185 



Abstract of Dissertation Presented to the Graduate School 
of the University of Florida in Partial Fulfillment of the 
Requirements for the Degree of Doctor of Philosophy 

EXPERIMENTAL AND THEORETICAL INVESTIGATION OF THE REACTIVITY 
OF PARTIALLY FLUORINATED RADICALS 

By 

Michael David Bartberger 

May 1998 

Chairman: William R. Dolbier, Jr. 
Major Department: Chemistry 

The reactivities of a series of partially-fluorinated radicals towards intermolecular 

addition, hydrogen abstraction, and intramolecular cyclization have been investigated. 

Based on competitive kinetic techniques and absolute rate contants for addition of these 

radicals to styrene obtained by laser flash photolysis, absolute rate constants for 

abstraction of hydrogen from tributylstannane have been determined for 1,1- 

difluoroalkyl, 2,2-difluoroalkyl, 1,1,2,2-tetrafluoroalkyl, 3-perfluoroalkyl, and 

pentafluoroethyl radicals. Fluorination at the 3-position of an alkyl radical was found to 

exert a negligible effect on the kinetics of hydrogen abstraction. All other systems 

exhibit rate enhancements relative to non-fluorinated analogues, the magnitudes of 

which are dependent upon the degree and location of fluorine substitution. A parallel 

computational study was performed utilizing density functional calculations, providing 

estimates of carbon-carbon and carbon-hydrogen bond dissociation energies (BDEs) for 

hydrofluorocarbons. The observed kinetic enhancements were attributed to a 

combination of structural, charge transfer, and enthalpic effects, due to the pyramidal 

nature of 1-fluoralkyl radicals, increased electrophilic character induced by successive 

fluorination, and thermodynamics of carbon-carbon and carbon-hydrogen bond 



VI 



formation. From the computation of partial atomic charges in fluoroalkanes, the 
contrasting effect of 1-fluorination on carbon-carbon and carbon-hydrogen BDEs and 
the consistent strengthening effect of such substitution at the 2-position have been 
explained on the basis of Coulombic interactions. 

Based on the rate constants obtained for hydrogen abstraction, absolute rate 
constants for 5-exo and 6-endo intramolecular cyclization for a series of partially 
fluorinated 5-hexenyl radicals have been obtained. These observed rates of cyclization 
may be rationalized by the same combination of effects influencing their bimolecular 
addition reactions. In some cases, the rate of 6-endo closure is dramatically 
accelerated relative to the parent hydrocarbon without the introduction of reversibility of 
ring closure. 



VII 



CHAPTER 1 
AN OVERVIEW OF ORGANIC FREE RADICAL REACTIONS 

Introduction 

The discovery of the first free radical, triphenylmethyl, by Moses Gomberg 1 in 
1900 initiated considerable effort directed toward the understanding of radical reactivity. 
However, only after a series of pioneering investigations undertaken more than thirty 
years later were the primary mechanistic pathways available to organic free radicals well 
elucidated. 2 " 6 These studies, most notably those of Kharasch et a/., 4 " 7 demonstrated that 
most radical processes can be expressed in terms of a small number of elementary 
steps, or variations thereof, as shown below in Figure 1-1. 8 



A* 


+ 


B' 




A-B 




coupling / homolysis 


(1-1) 


A* 


+ 


B-D 


■^^ 


A-B + 


D' 


substitution (Sh2) 


(1-2) 


A* 


+ 


B=D 


^^ 


A-B-D * 




addition / p-fission 


(1-3) 


A* 


+ 


e 


~^— 


A" ; 




electron transfer 


(1-4) 


A* 


_ 


e 


-i— 


A + 









Figure 1-1. The Elementary Mechanistic Pathways of Free Radical Reactions. 

Despite these breakthroughs, organic free radical reactions continued for years 
to be regarded as unpredictable, unselective, and in general inadaptable to synthetic 
application. Fortunately, subsequent kinetic and thermochemical studies have served to 
uncover the factors governing the reactivity of organic radicals, and consequently in 
recent years sentiment toward the utility of free radicals in synthesis has drastically 
changed. Indeed, the number of elegant works in the literature based on radical 



1 



mediated transformations is a testament to their applicability in the construction of 
natural products and other complex synthetic targets. 9 " 15 It is the purpose of this 
introductory chapter to acquaint the reader with the fundamental types of free radical 
processes which occur in organic systems, as well as to provide an overview of the 
wealth of physical studies which have given rise to the current level of understanding of 
organic radical reactivity. 

Radical Chain Processes 

Most free radical reactions occur via a sequence of chain events, propagated by 
intermediate steps during the course of the reaction. An example illustrating a 
competition between two potential pathways is provided in Figure 1-2. 

Initiation: In-ln *- 2 In* (1-5) 

In" + M-H ► In-H + M ' (1-6) 

Propagation: IvT + R-X *► M-X + R' (1-7) 

R- + M-H £_». r_h + M* (1-8) 

R* *- R" (1-9) 

Propagation: r- • + M-H ► R'-H + M- (1-10) 

Figure 1-2. Radical Chain Process Involving Competition Between Rearrangement 
versus Hydrogen Atom Transfer from a Donor Molecule M-H. 

Homolysis of an initiator, typically accomplished by thermal or photochemical 
means, provides a source of (often metal centered) radicals M* (equation 1-6) from 
which intermediate radicals R* are generated by reaction with a suitable precursor R-X 
(equation 1-7). This species encounters one of two fates: trapping, in this case by 
hydrogen atom donor, to yield R-H (equation 1-8) or transformation via a unimolecular or 
bimolecular process (equation 1-9) to form radical R", itself then trapped producing 



R'-H. In either case, additional metal radicals are formed and the chain process 
continued via the propagation steps given in equations 1-7, 1-8 and 1-10. 

The distribution of products R-H and R'-H is governed by the relative propensity 
of R' toward rearrangement versus trapping (that is, k t and k H ,) the latter dependent on 
both the nature of R* and the type of trapping agent employed. In systems where 
trapping is fast relative to rearrangement (k H » k r ), the partitioning radical R* is 
converted to R-H with little or no rearranged product. However, if k H and k r are of 
comparable magnitude, product mixtures result. An understanding of the reactivity of a 
radical intermediate toward potential competing processes is therefore essential for the 
design of useful kinetic experiments, as well as for the development of effective synthetic 
strategies. 

For an efficient chain process, it is necessary that the propagation steps are 
rapid relative to chain termination steps, thereby maintaining a low but constant 
concentration of radical intermediates. Besides the obvious practical benefit (higher 
product yields) resulting from such a condition, the occurrence of undesired chain 
termination side reactions such as disproportionation and radical-radical coupling, 
possibly complicating kinetic analyses, is minimized. In many cases, this may be 
achieved by judicious selection of the type and concentrations of precursor R-X and 
trapping agent. 

This procedure enjoys wide application in both kinetic and synthetic studies 
requiring the controlled generation of radical intermediates. One of its variants, likely the 
most commonly used procedure for the indirect (competitive) determination of the rates 
of organic radical reactions, is based on the trialkylstannane reduction of an alkyl halide 
(the "Tin Hydride Method"). 16 " 18 Other modifications of this general procedure exist, 
accommodating a variety of radical precursors and trapping agents; a discussion of time- 
resolved and competitive techniques utilized in radical kinetic measurements is provided 
later in the chapter. 






Finally, it is important to note the implication of kinetic control in the above 
discussion. That is, that the distribution of products R-H and R'-H may be ascribed to 
the relative values of k H and k r hinges on the absence of any thermodynamic 
equilibration of products under the reaction conditions. This is of vital importance in the 
design and interpretation of competitive kinetic studies and is discussed in detail in 
Chapter 3. 

Radical reactivity is dependent on the "complex interplay" of thermodynamic, 
steric, and polar considerations. 19 The relationship between enthalpies of activation and 
heats of reaction, the basis of the thermochemical kinetic approach of Benson, 20 was 
recognized early on and holds for a number of radical addition and substitution 
reactions, where the order of reactivity often parallels exothermicity. 21 " 23 This relation 
has led to such overgeneralizations as "radical reactions follow the most exothermic 
available pathway" or ". . . afford the most stable possible product." 8 However, reaction 
thermochemistry is not the sole, nor even predominant decisive factor in the outcome of 
radical reactions. Nonbonding interactions and the electronic influence of substituents in 
ground and transition states (which may be rationalized in terms of Frontier Molecular 
Orbital (FMO) theory) 24 25 will also play a role. A discussion of the combination of these 
effects as manifested in hydrogen atom abstractions and inter- and intramolecular 
additions, the most commonly occurring and well-characterized reactions of organic free 
radicals, will now be presented. 

Hydrogen Atom Abstraction Reactions 

The vast majority of free radical applications involve the use of an organometallic 
hydride of the type R 3 M-H (most commonly, where M = Sn, Si, or Ge) as a hydrogen 
atom donor and chain propagation agent, the properties of which have been the focus of 
extensive investigation by kineticists. Metal-hydrogen bond dissociation energies 
(BDEs) along with activation parameters and associated absolute rate constants for 



hydrogen atom transfer to n-alkyl hydrocarbon radicals by a series of donors R 3 M-H 
have been determined and are provided below in Table 1-1 , 16 ,26 ~ 32 



Table 1-1. Bond Dissociation Energies with Activation Parameters and Rate Constants 
for Hydrogen Atom Transfer to Hydrocarbon Radicals by R 3 M-H. 



R3M-H 


BDE, kcal 


mol 1 


loq A 


E a , 


kcal mol" 1 


kH, 


10 


3 M- 1 s" 1 (298 K) 


nBu 3 SnH 


73.7 




9.06 




3.65 






2.3 


(TMS) 3 SiH 


79.0 




8.86 




4.47 






0.38 


nBu 3 GeH 


82.6 




8.44 




4.70 






0.093 


(TMS) 2 Si(CH 3 )H 


82.9 




8.89 




5.98 






0.032 


Et 3 SiH 


90.1 




8.66 




7.98 






0.00064 



Analysis of the data demonstrates that for hydrogen atom abstraction by 
structurally similar radicals from this series of donors, a direct relation holds between the 
rate of transfer and the strength of the metal-hydrogen bond being broken. This is 
depicted graphically in Figure 1-3. In addition, it is noted that in each case the pre- 
exponential term in the Arrhenius relation remains relatively constant. Thus, the rate 
variations within the series are due almost entirely to differences in activation energies. 



16 
14 



3 12 

CNJ 



c 



10 



72 



m = -0.50403 
b =52.313 
r 2 = 0.96858 




BDE, kcal mol 



-1 



Figure 1-3. Plot of In k H for Alkyl Radicals versus M-H Bond Dissociation Energies for 
Hydrogen Atom Donors R 3 M-H in Table 1-1. 



6 

However, as previously mentioned, relative thermodynamics is not the only factor 
which influences the kinetics of hydrogen atom transfer. The fast donor thiophenol 
(PhSH) reacts with primary alkyl radicals with a rate constant of 1.36 x 10 8 M' 1 s" 1 at 
298 K, 33 and has been employed as a trapping agent in competitive kinetic studies 
involving strained or otherwise highly reactive radicals with rearrangement rates upward 
of 10 11 s" 1 and thus with lifetimes on the picosecond timescale. 34 This enhanced rate of 
transfer, not commensurate with its S-H BDE of 82.0 kcal mol" 1 , 35 gives rise to a severe 
deviation from the plot in Figure 1-3 and indicates the presence of other influences. 



Table 1-2. Absolute Rate Constants for Hydrogen Atom Transfer to terf-Butoxyl 
Radicals by R 3 M-H. 



R 3 M-H 


kin, 


10 6 


M- 1 s' 1 (300 K) 


nBu 3 SnH 




220 


(TMS) 3 SiH 






110 


nBu 3 GeH 






80 


Et 3 SiH 






5.7 



Further evidence may be found in the rates of hydrogen atom transfer to tert- 
butoxyl radicals by the same series of hydrogen atom donors, provided in Table 1-2 and 
illustrated graphically in Figure 1-4. 3637 It is observed that for fe/f-butoxyl radicals, rates 
of hydrogen abstraction are at least two orders of magnitude greater than those of their 
n-alkyl counterparts. Although the relative strengths of the newly formed C-H or O-H 
bonds will certainly play a role, the difference in BDE between rBuO-H and n-alkyl C-H 
bonds (105 and 100 kcal mol" 1 , respectively) 38 is not sufficient to explain the increase in 
reactivity, especially in light of the fact that such rapid hydrogen atom abstractions 
should proceed with early transition states. 36 

At this time, the absolute rates of reduction of ferf-butoxyl radicals by thiophenol 
have yet to be determined. However, a series of competition studies by Hartung and 



Gallou 39 involving 4-pentenyl-1-oxy radicals and utilizing naphthalene 2-thiol (NpSH) as 
a trapping agent have determined a ratio [ k H (NpSH) / k H (nBu 3 SnH) ] of 1.4. By 
comparison, n-alkyl radicals afford the ratio [ k H (PhSH) / k H (nBu 3 SnH) ] = 59.1. 
Although a leveling effect may be partly responsible for the compressed ratio of rates for 
tert-butoxyl radicals (which are indeed within an order of magnitude of the diffusion- 
controlled limit) 40 it is logical to assume based on the aforementioned examples that 
hydrogen abstraction reactivity will be governed to some extent by factors other than 
simple relative BDE values of the donor species. 





19 - 


• ^"^--^v. 






m = -0.22212 
b =35.942 


* 






^"~-\» 




r 2 = 0.91532 


o 
o 

CO 


18 - 






• 




I 


17 - 










c 


16 - 
15 - 








• ^^*< 




I I 


I I 


I I 


I I I 



72 74 



76 78 



80 82 
BDE, kcal ■ mol 



84 

1 



86 88 90 92 



Figure 1-4. Plot of In k H for re/t-Butoxyl Radicals vs. M-H Bond Dissociation Energies for 
Hydrogen Atom Donors R 3 M-H in Table 1-2. 



Chatgilialoglu er a/. 3637 have attributed such differences in reactivity to a 
polarized, or charge separated, transition state of the type depicted in Figure 1-5. Here 
it can be seen that in the case of an electropositive metal hydride donor, hydrogen atom 
transfer to alkoxyl radicals (b) is facilitated by greater stabilization of partial negative 
charge on oxygen relative to carbon, with a resultant decrease in activation barrier. 



5 + 8" 

(a) R 3 M H R 



5 + 5" 

(b) R,M H OR 



Figure 1-5. Charge Polarized Transition State for Hydrogen Abstraction from R 3 M-H by 
(a) Alkyl and (b) Alkoxyl radicals. 



8 

In the case of thiol donors, the opposite situation ensues. The greater 
electronegativity of sulfur relative to tin (or other metal atom) gives rise to a reversal in 
the transition state charge distribution. This arrangement, involving a partially negatively 
charged sulfur atom, is better suited to the more nucleophilic alkyl radical, where in the 
alkoxyl case a less- or non-polarized transition state results. Such a mismatch in the 
latter is partially responsible for the decrease in rate enhancement for hydrogen 
abstraction from thiols by alkoxyl radicals, relative to their alkyl analogues. 

Frontier Molecular Orbital Theory of Atom Abstraction Reactions 

FMO theory provides a satisfying rationale for the kinetic characteristics of 
hydrogen abstraction reactions of free radicals. In general terms, radicals are species 
containing an unpaired electron in a singly occupied molecular orbital (SOMO), which in 
the ground state of the radical is its highest occupied orbital. According to the FMO 
concept, during the course of the reaction this SOMO will interact with both the highest 
occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals of the donor 
molecule. Such interactions between these "frontier molecular orbitals" 24 are not 
necessarily equal. The extent of SOMO-HOMO and SOMO-LUMO interaction is 
governed by their energy values, the strongest interaction occurring between orbitals 
closest in energy. 

It is these values, influenced by atom type as well as neighboring substituents, 
from which the relative descriptors such as "nucleophilic" and "electrophilic" are derived 
and provide the basis for the previously described concept of "polar factors." Electron 
donating substituents generally serve to raise both HOMO and LUMO energies, with 
electron withdrawing groups resulting in lowering. Radicals possessing a low energy 
SOMO will display electrophilic character, whereas a higher energy SOMO gives rise to 
a more nucleophilic species. Figure 1-6 depicts the FMO interactions between radical 
and donor in each of these cases. 



9 

During the abstraction process, the primary interaction involves the radical 
SOMO and the o and o* orbitals of the donor M-H bond. The antibonding o* orbitals of 
the donor are typically quite high in energy, and thus in atom abstraction reactions the 
SOMO-HOMO interaction dominates. 



SOMO 

1 \ > 

> > 

SOMO 
HOMO 



(a) (b) 



Mf 



HOMC 



HP 



Figure 1-6. FMO Diagram Illustrating the SOMO-HOMO Interaction Between (a) a 
Nucleophilic Alkyl and (b) an Electrophilic Alkoxyl radical. 



Here it is seen that the lower-energy SOMO of the alkoxyl radical (ca. -12 eV, as 
determined from ionization potential measurements) 25 leads to a reinforced interaction 
with the donor HOMO (case b). This greater stabilizing interaction results in a lowered 
activation barrier and hence a more facile transfer reaction, compared to the more 
nucleophilic alkyl radical, (case a) whose SOMO energies range from -6.9 to -9.8 eV. 25 

Some of the most striking examples of such "polar" factors involve systems 
where fluorine substitution has taken place at, or adjacent to, the radical center. This is 
elaborated upon in Chapters 2 and 3, where the effects of fluorination on the structure 
and reactivity of free radicals are discussed and compelling evidence provided based on 
kinetic studies of hydrogen transfer to such partially and fully fluorinated alkyl radicals. 

Intermolecular Radical Addition Reactions 

Over the past twenty years, the intermolecular addition reactions of free radicals 
(as well as their intramolecular cyclization counterparts) have become an important 



10 

addition to the arsenal of C-C bond formation methods available to the synthetic organic 
chemist. Their mild means of generation from a variety of precursors and tolerance for a 
wide variety of functional groups provide distinct advantages over ionic processes. 

Alkyl radical additions to carbon-carbon double bonds are highly exothermic, as a 
new o bond is formed at the expense of a ft bond (in the case of methyl radical addition 
to ethylene (Figure 1-7), AH ren = ca. -22.6 kcal mol" 1 ). 4142 In accordance with the 
Hammond postulate, 43 such additions should proceed via early transition states, with low 
barriers of activation. This is indeed the case, as supported by a wealth of both 
experimental 44 and theoretical 41 ,45 ~ 49 data. 

CH 3 " + CH 2 =CH 2 ** CH 3 CH 2 CH 2 * 

E a = 7.9 kcal mol" 1 
AHrxn = -22.6 kcal mol" 1 

Figure 1-7. Addition of Methyl Radical to Ethylene, Yielding n-Propyl Radical. 
Experimental Activation Energies and Heats of Reaction are Shown. 

FMO Theory of Radical Additions 

Quantum mechanical molecular orbital calculations at levels of ab initio theory 
ranging from UHF to UQCISD(T) and varying basis sizes from 3-21G to 6-311G(2df,p) 
are consistent in their characterization of the transition structure for the above reaction 




7 Vr(CH 3 ) = 101.9° 

\ 
\ 

2.246 A \ 109.1° 



173.4° 






1.382 A 
154.7° 

Figure 1-8. UHF/6-31G(d) Transition Structure and Relevant Geometrical Parameters 
for Addition of Methyl Radical to Ethylene. 



11 



LUMO 



LUMC 



SOMO 



rh 



4 



HOMO 



(a) 



SOMO 



(b) 



,Hf 



HOMC 



Figure 1-9. FMO Diagram for Addition of (a) Nucleophilic and (b) Electrophilic Radicals 
to Alkenes. 



Table 1-3. Some Relative Rates of Addition of Methyl and tert-Butyl Radicals to Alkenes 
CH 2 =CHX. 



X 


k rP | (CHa*) 


krel 


«CH 3 ) 3 C') 


H 


1 




1 


CH 3 


0.7 




0.74 


OCH 2 CH 3 






0.31 


F 


0.9 






CI 


4.2 




13.2 


CN 


34 a 




1920 



Data for ethyl radical. 



(Figure 1-8), which possesses an incipient C-C bond distance of ca. 2.23 - 2.27 A. The 
C-C-C attack angle of 109.1° is rationalized in FMO terms based on a primary interaction 
between the radical SOMO and the LUMO of the alkene. It is in such reactions with high 
exothermicities and early transition states that FMO interactions are most 
substantial. 2450 This postulate enjoys experimental support; for the Nbutyl radical, a 
correlation exists between rates of addition to alkenes and the experimentally 



12 

determined electron affinities of the latter. 21 51 Such an FMO interaction for nucleophilic 
radicals is shown in Figure 1-9 (case a) and is influenced by substituents on both radical 
and olefin, a raising of radical SOMO and/or lowering of alkene MO energies 
strengthening the SOMO-LUMO interaction and enhancing the rate of addition as seen 
from the data in Table 1-3. 21 52 Here, the greater nucleophilic character of Nbutyl relative 
to methyl is evident from its enhanced rate of addition to olefins bearing electron 
withdrawing groups. 

As previously discussed, electronegative substituents at the radical center which 
substantially lower its SOMO energy will impart electrophilic character and reinforce the 
transition state SOMO-HOMO interaction (Figure 1-9, case b). Indeed, it has been 
shown that rates of addition of dicyanomethyl 53 and perfluorinated 54 radicals correlate 
with the ionization potentials of the substrate alkenes. The intermediate behavior of 
"ambiphilic" radicals, such as malonyl and (tert-butoxycarbonyl)methyl, has also been 
documented, these species yielding "U"-shaped correlations between rates of addition 
and alkene IP and EA values. 55 " 57 

A more thorough presentation of kinetic results obtained to date for the addition 
of partially and fully fluorinated radicals to alkenes is given in Chapters 2 and 3. 

Steric Effects; Regiochemical Preferences in Addition 

Competition studies on both nucleophilic and electrophilic radicals have provided 
for some generalizations in terms of the regiochemical preference for addition to 
unsymmetrically substituted olefins. 52 The preferred orientation of radical addition 
occurs to the unsubstituted end of the double bond, attributed to steric repulsion but also 
influenced by the effect of substituents on the coefficients of the HOMO and LUMO of 
the alkene. Such FMO effects can be the decisive factor in polysubstituted olefins if 
steric effects are in opposition. Strongly spin delocalizing substituents on the alkene 
reinforce such sterically induced regiochemical preferences and exert slight rate 



13 

enhancing effects; however, as such additions occur through early transition states the 
effect of exothermicity on the kinetics of addition should be minimal. 

The concepts introduced in the aforementioned discussion on intermolecular 
radical additions extend to their intramolecular cyclization analogues, an overview of 
which will now be presented. 

Intramolecular Addition Reactions: Radical Cyclizations 

The intramolecular addition reactions of alkenyl radicals enjoy a strong foothold 
among the available strategies for the construction of cyclic organic molecules. In 
addition to their synthetic utility, the kinetic, regioelectronic and stereoelectronic 
characteristics of radical cyclization reactions as a function of substituent continue to fuel 
an abundance of fundamental physical organic structure-reactivity investigations, more 
than thirty years after the first report of the archetypal radical ring closure, cyclization of 
hex-5-en-1-yl radical 1 (Figure 1-10). 58 



25° C 





(98%) (2%) 

1 2 3 

Figure 1-10. Cyclization of Hex-5-en-1-yl Radical 1 to Cyclopentylcarbinyl (2) and 
Cyclohexyl (3) Radicals. At 25° C, 5-exo Closure Dominates 49 : 1. 

The most striking aspect of this reaction lies in the preferred regiochemistry of 
addition. In the case of the parent hydrocarbon, 5-exo 59 cyclization dominates (E ac t [5- 
exo] = 6.8 kcal mol" 1 , E act [6-endo] = 8.5 kcal mol" 1 ) 2760 yielding the less 
thermodynamically stable primary cyclopentylcarbinyl radical 2. This finding has 
provided the driving force for a number of experimental 61 and theoretical 62 " 68 



14 

investigations geared toward the understanding of radical cyclization regiochemistry in 
the hydrocarbon and related substituted systems. 

Early explanations, 69 later advanced by semiempirical techniques, 65 attributed 
this result to a less negative entropy of activation for cyclization to 2. Although 
experiment demonstrates this to be true, the difference (AS* 15 - AS* 1i6 = 2.8 eu) 60 is 
insufficient to completely account for the observed regiochemistry; the preferred mode of 
cyclization resulting primarily from enthalpic (AH* 16 - AH* 15 = 1.7 kcal mol" 1 ) rather than 
entropic factors. Ab initio computations 66 lend support to this conclusion. 

Transition structures for 5-exo and 6-endo cyclization of 1 have been located 
using a variety of theoretical treatments. The UHF/6-31G(d) structures leading to 2 and 
3 are shown in Figures 1-11 and 1-12. 




c=&^4 



9 = 109.7° 

2.186A 



of* 




30 



Figure 1-11. Two Views of the UHF/6-31G(d) "5-exo-chair" Transition Structure for 
Cyclization Hex-5-en-1-yl Radical 1 to Cyclopentylcarbinyl Radical 2. 




9 = 98.4° 
2.260 A < 




Figure 1-12. Two Views of the UHF/6-31G(d) "6-endo-chair" Transition Structure for 
Cyclization of Hex-5-en-1-yl Radical 1 to Cyclohexyl radical 3. 



15 

Spellmeyer and Houk 66 have also postulated additional "boat-like" transition 
structures on the basis of molecular mechanics calculations parameterized by ab initio 
results of model systems. Inclusion of these "boat-like" structures as viable competing 
pathways was found to be necessary for the accurate prediction of regio- and 
stereoselectivities in cyclizations of alkyl substituted and heteroalkenyl radicals. The 
existence of such structures is corroborated by higher level ab initio treatments 
performed as part of the present study and are shown in Figures 1-13 and 1-14. 





Figure 1-13. Two Views of the UHF/6-31G(d) "5-exo-boat Transition Structure for 
Cyclization of Hex-5-en-1-yl Radical 1 to Cyclopentylcarbinyl Radical 2. 





Figure 1-14. Two Views of the UHF/6-31G(d) "6-endo-twist-boaf Transition Structure 
for Cyclization of Hex-5-en-1-yl Radical 1 to Cyclohexyl Radical 3. 



Upon inspection of the forming bond lengths and angles of the 5-exo structure in 
Figure 1-11, its similarity to the transition structure for addition of methyl radical to 
ethylene is readily apparent. The C-C-C angle of attack, 109.7°, is practically identical to 



16 

that in Figure 1-8 and fits the requirement for overlap of the radical SOMO with the n* 
orbital of the alkene moiety. From Figure 1-12 it is observed that this angle is 
significantly reduced (98.4°) in the 6-endo approach. Thus the required disposition of 
centers for optimal FMO overlap is more readily achieved in the 5-exo transition 
structure, leading to the kinetically preferred cyclization product. 

Substituent Effects on the Kinetics and Reqiochemistry of 5-Hexenyl Cyclizations 

Radical 1 cyclizes to cyclopentylmethyl radical 2 with a rate constant (/c C 5) of 
2.3 x 10 5 s" 1 at 25 °C. 27 In the parent hydrocarbon, 6-endo closure competes to a very 
minor extent (k ce = ca. 4.7 x 10 3 s' 1 ). However, it will be shown that the rates and 
regiochemical preferences can be substantially affected by substitution at both the 
radical center and terminal alkene. 

Alkyl Substitution; Steric Effects 

Beckwith et al. have reported rate constants for a number of alkyl-substituted 
hexenyl radicals. 70 71 The rates of 5-exo closure as a function of gem-dialkyl substitution 
on the aliphatic portion of the hexenyl chain is shown in Table 1-4. 

It is seen that substitution at the radical center has a nearly negligible effect on 
the rate of cyclization, due to offsetting polar and steric considerations. Conversely, a 
significant (> 10-fold) rate enhancement is observed with internal substitution (systems 
6, 8, and 9), accelerated by relief of steric compression between alkyl groups during ring 
formation (the "Thorpe-lngold", or "gem-dimethyl" effect). 72 

Intermediate kinetic behavior would be expected of monosubstituted 5-hexenyl 
systems. The data in Table 1-5 show this to be the case. In addition, a stereochemical 
preference for cis- or frans-dimethylcyclopentanes, depending on the location of the 
substituent on the chain, is observed. This has been rationalized by Beckwith et a/. 71 






17 






Table 1-4. 5-exo Cyclization Rate Constants for gem-Disubstituted 5-Hexenyl Radical 
Derivatives. 



Cyclization Reaction 



k C5 , 10 5 s" 1 (298 K) 




2.3 C 





3.0 £ 



^ 




36 c 




8 




52 c 



^ 




32 c 



Reference 27. b Reference 28. c Reference 71 . 



based on the cyclization transition structure depicted in Figure 1-11. Although "early" in 
terms of the forming C-C bond, the overall orientation of atoms in this structure is quite 
product-like. According to this rationale, substituents on the aliphatic fragment are 
likened to those in chair cyclohexane, which then occupy an equatorial position in the 
chair transition structure. Minor products are assumed to derive from the occupation of 



18 

axial positions. The latter has been disputed by Spellmeyer and Houk, 66 whose model 
indicates that such secondary products originate from an equatorial disposition of 
substituents in the boat-like transition structure of Figure1-12, rather than from an axial 
orientation in the chair. 



Table 1-5. Cis- and Trans- 5-exo Cyclization Rate Constants for Monosubstituted 5- 
Hexenyl Systems. Rate Constants are for 298 K. 

Cyclization Reaction k C5 (cis), 10 5 s" 1 k C5 (trans), 10 5 s" 1 





1.1 8 0.42 £ 



10 c/s-11 frans-11 





2A b 4.5 b 



12 c/s-13 frans-13 





7.0 b 2A b 



14 c/s-13 frans-13 





,\\ 



0.75" 3.6 b 



15 c/s-11 trans- 11 

3 Reference 73. " Reference 71 . 

In the above examples, 5-exo products are formed either predominantly (> 97%) 
or exclusively. Substitution at the vinyl group leads to marked changes in regiochemical 
ratios as indicated by the data in Table 1-6. Replacement of hydrogen by methyl (16) or 
isopropyl (19) at C 5 results in preferential formation 6-membered rings 18 and 21. 
Inspection of the data indicates that this shift in regiochemistry is not due to a significant 



19 



Table 1-6. Rate Constants at 338 K for 5-exo and 6-endo Cyclization for Vinyl- 
substituted 5-Hexenyl Radicals. 



Cyclization Reaction 



*cs. 10 5 s" 1 k C6l 10 5 s" 1 





16 




19 




22 





3 





17 



18 





20 



21 





23 



24 





9.4 C 



0.21' 



0.21' 



1.0 C 



22 £ 



0.19 



a, b 



0.37 £ 



0.66 c 



0.1 9 C 



<o.r 



25 26 27 

Calculated from the Arrhenius parameters given in reference 27. ' Reference 70. 



extent to rate enhancement for 6-endo closure, but rather a substantial (44-fold) 
retardation of 5-membered ring formation due to a combination of 1,5 steric hindrance 
and back strain engendered at C 5 upon adaptation of sp 3 character. Substitution at both 
C 5 and C 6 again favors 5-exo closure, the rate of which decreased relative to the parent 
system. Disubstitution at C 6 (25) gives rise to a slight (2.3-fold) rate enhancement for 5- 
exo closure, sufficiently explained on thermodynamic grounds, which dominates 6-endo 



20 

cyclization by a factor of at least 220. The above data indicates that the kinetic and 
regiochemical characteristics of alkyl-substituted 5-hexenyl cyclizations may be 
sufficiently rationalized by steric considerations. The effects of substitution by 
conjugating, heteroatom-containing groups is outlined below. 

FMO Considerations 

Studies of 5-hexenyl systems bearing "polar" subsitutents have been 
investigated by Newcomb. 1674 " 77 In line with those of intermolecular radical additions, 
the kinetics of radical ring closure will be influenced by the impact of substitutents on the 
SOMO-HOMO and SOMO-LUMO interaction in the cyclization transition state. 

Table 1-7. 5-exo Cyclization Rate Constants For a-Donor- and a-Acceptor-substituted 
6,6-Diphenyl-5-hexenyl radicals. Rate Constants are for 298 K. 

X Y 



^* • 




5-exo 


X Y Ph 


^^^^\ 


^^ 


Cr Ph 




Ph 






System 


X 


Y 


k<*. 10 7 s 1 (298K) a 


28->29 


H 


H 


4 


30->31 


H 


CH 3 


2 


32->33 


CH 3 


CH 3 


1 


34->35 


H 


OCH 3 


4 


36->37 


H 


C02Cn2CH 3 


*3.7 


38->39 


CH 3 


C02CH2CH 3 


0.04 


40->41 


CH 3 


CN 


0.03 



a Calculated from the Arrhenius parameters provided in reference 76. 

As seen from the data in Table 1-7, only a very minor effect is exerted by either 
a-donor or a-acceptor substituents, relative to parent system 28. The marked decrease 



21 

in rate for 38 and 40 is attributed to an increase in activation energy due to enforced 
planarity at the radical site induced by the 7i-delocalizing substituents C0 2 CH 2 CH 3 and 

CN 51,78 



Table 1-8. Absolute Rate Constants at 298 K for 5-exo Cyclization of 5-Hexenyl 
Systems Bearing Vinylic Donor and Acceptor Substituents. 

Cyclization Reaction /c C5 , 10 5 s" 1 (298 K) 





H 3 CO- _ 42 „ 




2.3 C 



42 43 

>CN 



NC"^i ; 



H3CO 





680 6 



1000 ft 



46 47 

a Reference 27. b Calculated from the Arrhenius parameters provided in reference 74. 

Substitution at the vinyl terminus, especially by strong resonance-withdrawing 
groups can significantly accelerate the rate of ring closure. Although possessing a 
radical stabilizing group, methoxy analog 42 enjoys only a very minor increase in rate. 
The donor substituent raises the energies of the frontier orbitals, increasing the SOMO- 
HOMO interaction but widening the SOMO-LUMO energy gap, the latter more important 



22 

for relatively nucleophilic alkyl radicals (Figure 1-9). Substituents which serve to lower 
the FMO energies should reinforce the SOMO-LUMO interaction, leading to rate 
enhancement. The nearly 300-fold increase resulting from cyano substitution (44) 
reflects such an effect. The slight rate increase 46 relative to 45 has been explained on 
the basis of the suggested slight extra "push-pull," or "captodative" stabilization 
manifested in donor-acceptor disubstituted systems. 79,80 

The importance of kinetic control was previously mentioned. Care must be taken 
in assessing the potential for reversibility in such intramolecular additions, which may 
obscure the effect of steric and/or polar influences on reaction kinetics. This is 
demonstrated in the 5-exo:6-endo product ratios of highly stabilized systems 48 and 

51 81,82 



Table 1-9. Product Ratios for 5-Hexenyl Cyclization Reactions Under Full or Partial 
Thermodyamic Control. 

Cyclization Reaction % 5- exo % 6-endo 




< 22 > 78 



48 49 50 

^NC^COjEt I CN ^J 

U — * OWt + k>° 2 Et 16 84 

51 52 53 

Given the data and discussion provided in the above sections, a review of the 
direct and competitive techniques utilized in the determination of rates of organic radical 
reactions is now in order. 

Methods for Determination of Organic Radical Kinetics 

The development of indirect competitive methods, in conjunction with laser flash 
photolytic generation and time-resolved detection of transient intermediates, has greatly 



23 

expanded the dynamic range available for the measurement of radical reactions, 
especially those at the upper end of the kinetic scale. Such advances have provided for 
the use of a variety of precursors and the accurate determination of rate constants for 
reactions approaching the diffusion-controlled limit in solution. 16 

Laser Flash Photolysis; Direct Measurement of Addition Rates 

In the time-resolved laser flash photolysis method, described in detail in the 
literature, 83 radicals are generated from precursors possessing a suitable chromophore 
by a laser pulse of the appropriate wavelength. Alkyl iodides, diacyl peroxides, and the 
O-acylthiohydroxamic esters of Barton et a/. 84 are most commonly utilized in this regard 
(Figure 1-15). 



(a) 


hv 
r? i 




R • 
hv 












rc-i 


rV Y R 









-C0 2 


2 


R 


(L>) 






fast 























XQ 




hv 




fAo' 


-co 2 


-^- R • 






(<0 


^* 


fast 






S 

















Figure 1-15. Laser Flash Photolytic Generation of Radicals R* from (a) Iodide, (b) Diacyl 
Peroxide, and (c) O-Acylthiohydroxamic Ester Precursors. 



These radicals so generated undergo further reaction, usually bimolecular 
addition or unimolecular cyclization to a (typically phenyl-substituted) double bond (a 
styrene in the case of bimolecular additions). The increase in the characteristic 
absorption of this intermediate benzyl radical (X max * 320 nm) is then followed in a time- 
dependent manner by UV-visible spectroscopy (Figure 1-16). For bimolecular additions, 
this experimental growth curve is fit to the expression in Equation 1-11, yielding absolute 



24 

rate of addition k aM . With such data in hand, this addition can now serve as a competing 
basis reaction for the determination of rates of other transformations involving the same 
or structurally similar radical. 

/fobs = k + /Cadd [alkene] (1-11) 



"&o 




R 
V 


(monitor) 


X max ca. 320 nm 


/fadd 





hv 
R-X *• 



Figure 1-16. LFP Generation of R* and Detection of Transient Benzyl Radical Adduct 
for Determination of Absolute Rate of Addition k aM . 



Indirect Methods (Competitive Techniques) 

Indirect kinetic methods involve the partitioning of an intermediate between two 
competing pathways, one with a known rate constant and the other whose rate constant 
is to be determined. Post facto product analyses, typically by chromatographic or 
spectroscopic means, provide a ratio of rate constants from which the new kinetic value 
is obtained. Such radical kinetic measurements usually involve competition between two 
bimolecular reactions or a bimolecular reaction competing with unimolecular 
rearrangement; examples of both instances are provided in Figures 1-17 and 1-18. 

Determination of rates of hydrogen abstraction by this method involves the 
generation of R* in the presence of two trapping agents; in Figure 1-17, styrene and the 
hydrogen atom donor. Both traps are typically present in excess to ensure pseudo-first 
order behavior. Radical R* may undergo addition to styrene (with known rate constant 
/(add) forming the intermediate benzyl radical, itself trapped with excess hydrogen donor, 
yielding closed shell product with a rate which is kinetically unimportant provided the 
addition reaction is irreversible. Alternatively, R* is trapped directly by hydrogen atom 
donor with a rate constant k H , which may be obtained from the pseudo-first order relation 



25 



[ reduced ] 
[ adduct ] 



[fr H ][R.][M-H] 



[fraddl [R«][CH 2 =CHR'] 



(1-12) 



where [reduced] and [adduct] are the final product concentrations, [M-H] is the 
concentration of hydrogen atom donor, and [CH 2 =CHR'] the concentration of alkene trap. 
For accurate kinetic determinations, a series of runs is performed where trap 
concentrations are varied, (again, maintaining at least a five-fold excess) a plot of 
product ratios versus that of trapping agents yielding the ratio k H I k zM . 



R 



O (vary) 
^add 



M-H (vary) 




R 
M-H "\ 



(A. 



kH 



RH 



Figure 1-17. Competition Between Bimolecular Addition to an Alkene with Rate 
Constant k aM and Bimolecular Trapping by Hydrogen Atom Donor with Rate Constant 



Determination of rates of cyclization are performed in a similar manner, using 
hydrogen atom abstraction (with the known value of k H ) as the competitive basis 
reaction. Unimolecular rearrangement competes with bimolecular trapping with an 
excess of hydrogen atom donor (Figure 1-18) to yield intermediate cyclic radicals, further 
trapped to form characterizable products. 

The ratio of products of cyclization versus hydrogen abstraction are obtained 
from the pseudo-first order relation in Equation 1-13, a plot of the ratio of products of 
hydrogen abstraction to those of cyclization as a function of trapping agent concentration 
affording ratios k C5 1 k H and k C6 1 k H . 

Finally, the importance of an efficient chain process should be reemphasized. 
High conversions of precursors, although important for any radical reaction, are crucial in 






26 

competitive kinetic experiments. The reliability of data resulting from indirect methods is 
directly dependent on high "mass balance" values, those of 90% or greater typically 
being desired. 



[ reduced ] 
[ cyclized ] 



[k H ] [r«] [M-H] 
[kcn][R'] 



(1-13) 



M-H (vary) 



RH 



^C5 



<C6 



M-H 




M-H 




Figure 1-18. Competition Between Bimolecular Hydrogen Atom Abstraction with Rate 
Constant k aM and Unimolecular 5-exo and 6-endo Cyclization with Rate Constants k C5 
and k C6 . 



Such competitive processes have been employed extensively by the Dolbier 
research group in the investigation of the rates of addition, hydrogen atom abstraction, 
and cyclization of a variety of fluorinated open shell systems, providing the first 
quantitative kinetic data for this class of reactive intermediates. 5485 " 91 

Conclusion 



The preceeding discussions have attempted to provide the reader with an 
introduction to the chemistry of organic free radicals. Kinetic data for hydrogen 
abstraction, addition, and cyclization reactions of hydrocarbon radicals, important 
benchmarks for comparison of reactivity with other substituted systems, was provided. 



27 

Substituent effects, rationalized on the basis of a combination of thermodynamic, steric 
and FMO considerations, were discussed. 

The following chapter provides a review of the effects of fluorine substitution in 
organic molecules, including fluorinated radicals. Previous research efforts in this area 
by the Dolbier group are summarized, setting the stage for the presentation of results of 
the current study. 









CHAPTER 2 
THE FLUORINE SUBSTITUENT IN ORGANIC SYSTEMS 

Introduction 

Incorporation of fluorine into organic molecules often imparts dramatic alterations 
in structure and reactivity. These effects are induced by three major characteristics 
inherent to the fluorine atom: extreme electronegativity, non-bonded electron pairs, and 
relatively small size. 

Fluorine possesses the highest electronegativity of all the elements, with a value 
of 4.10 on the Pauling scale, compared to oxygen (3.50), chlorine (2.83), bromine (2.74), 
carbon (2.50), and hydrogen (2.20). 92 As a substituent in organic systems, this results in 
strong inductive withdrawal of electron density through the a molecular framework and 
highly polarized bonds with substantial ionic character. 

Three non-bonding pairs of electrons in 2p orbitals similar in size to those of 
other second-row elements provide for optimal overlap, and therefore an offsetting back 
donation of electron density into the molecule to which it is bonded. 

The accepted van der Waals radius of fluorine, 1.47 A, suggests minimal steric 
impact in comparison with other halogens (chlorine, 1.73 A; bromine, 1.84 A; iodine, 
2.01 A; carbon, 1.70 A; oxygen, 1.52 A; hydrogen, 1.20 A). 93 This has allowed for the 
complete replacement of hydrogen by fluorine in organic systems, a feat not possible to 
such an extent with any other element. 

The following sections, based on a number of excellent reviews, 9497 provide an 
introduction to the fascinating behavior exhibited by fluorinated stable molecules and 
reactive intermediates due to a combination of the above effects. 



28 



29 
Structure, Bonding, and Reactivity in Saturated Systems 

The data provided in Tables 2-1 and 2-2 9597 reveal a trend unique to fluorine 
within the halogenated methanes. An incremental shortening of C-F interatomic 
distances, with a resultant increase in bond dissociation energies, is observed as the 
series is traversed. No such trend exists for any other member of the halomethane 
family; on the contrary, it is seen from the data in Table 2-2 that such C-X BDE values 
instead decrease with increasing halogen content. Strengthening of C-H bonds is also 
observed within the fluoromethanes (CH 3 F, 101.3 kcal mol -1 ; CH 2 F 2 , 103.2 kcal mol" 1 ; 
CF 3 H, 106.7 kcal mol 1 ). 95 

Table 2-1. Carbon-Halogen Interatomic Distances (Angstroms) of Halomethanes 

X CHgX CH ?X? CHX3 CX4 

F 1.385 1.357 1.332 1.319 
CI 1.781 1.772 1.758 1.767 
Br 1.939 1.934 1.930 1.942 



Table 2-2. Carbon-Halogen Bond Dissociation Energies (D°, kcal mol 1 ) of 
Halomethanes 

CX 4 



72.9 
56.2 



Data for geminally fluorinated ethanes parallel that of the methane series, 
demonstrating a progressive strengthening and shortening of both C-C and C-F bonds 
with increasing fluorination (Table 2-3). Conversely, vicinal fluorination gives rise to the 
opposite effect on C-C bonds, a steady lengthening and weakening being observed. 



X 


CH3X 


CH2X2 


CHX 


F 


108.3 


119.5 


127.J 


CI 


82.9 


81.0 


77.7 


Br 


69.6 


64 


62 


I 


57.2 


51.3 


45.7 



30 

A variety of hypotheses have been put forth to explain the observed trends. One 
rationalization, invoked by Pauling 98 and based on valence bond theory, involves "double 
bond, no bond" resonance of the type depicted in Figure 2-1. 

Table 2-3. Interatomic Distances and Dissociation Energies of Fluoroethanes 

Ethane r (C-C), A D° (C-C). kcal mol' 1 r (C-F), A D° (C-F). kcal mol' 1 



CH3-CH3 


1.532 


90.4 


- 


- 


CH3-CH2F 


1.502 


91.2 


1.398 


107.9 


CH3-CHF2 


1.498 


95.6 


1.343 


Unknown 


CH3-CF3 


1.494 


101.2 


1.335 


124.8 


CH 2 F-CF 3 


1.501 


94.6 


Unknown 


109.4 (CH 2 F) 


CF3-CF3 


1.545 


98.7 


Unknown 


126.8 



As the degree of geminal fluorination is increased, the number of such resonance 
forms involving doubly bonded fluorine increases (0, 2, 6, and 12 in the case of CH 3 F, 
CH 2 F 2 , CH 3 F, and CF 4 , respectively). This is supported by ab initio calculations at the 
RHF/4-31G and 4-31 G(d) levels, 99 " 101 which illustrate back donation of electron density 
from fluorine into the C-F a* orbitals. It is further observed that the overlap population 
between the 2p orbitals of carbon and those of fluorine increases continually with 
successive fluorination; in contrast, such carbon-chlorine overlap populations decrease 
steadily from CH 3 CI to CCI 4 . 



F F 

= F 



-- w u... 



Figure 2-1. "Double Bond, No Bond" Resonance in Geminally Fluorinated Alkanes. 

Alternative explanations based on hybridization schemes have also been 
advanced. It is postulated that for electronegative elements bound to carbon, 
rehybridization occurs causing an increase in the amount of p character directed toward 



31 

the substituent. Thus, in CH 3 F, the C-F bond possesses greater p character, with 
greater s character in the C-H bonds. This rationale accounts not only for incremental 
C-F bond strengthening, but also for the observed changes in geometry within the 
fluoromethane series. Accumulation of p character in C-F bonds should lead to a 
decrease in FCF bond angle, accompanied by HCH widening. This is consistent with 
experimental observation (ZFCF in CF 4 , 109.5°; CHF 3 , 108.7°; CH 2 F 2 , 108.3 ; 102103 for 
CH 2 F 2 , ZHCH = 113.7°). 103 

Finally, a more recent argument has been advanced by Wiberg, 104 on the basis 
of Coulombic interactions between carbon and fluorine substituents. From charge-fitting 
treatments based on calculated electrostatic potentials, a linear increase in positive 
charge on carbon is observed, while the degree of negative charge on each of the 
fluorine substituents remains quite constant. Thus, incremental fluorine substitution 
strengthens not only new, but also previous, C-F bonds. This finding will be further 
discussed in Chapter 3, as such ESP-derived charges calculated for larger fluorinated 
systems as part of the present study were found to provide a cogent explanation for the 
remarkable and contrasting effects of fluorination on the strengths of both C-C and a and 
p C-H bonds. 

Structure. Bonding, and Reactivity in Unsaturated Systems 

The most reliable structural and tt-BDE data for the fluoroethylenes is provided in 
Table 2-4. 9597 Vinylic fluorine substitution results in shorter C=C bonds than in the 
parent hydrocarbon, and shorter C-F bonds than fluoroalkanes bearing the same 
number of geminal or vicinal fluorine substituents. Ab initio investigations by Radom" 
and others 105 " 107 attribute the C-F bond contraction to derealization of fluorine 2p 
electrons into the C-C n bond (depicted in resonance terms in Figure 2-2). Computed 
atomic charges are consistent with this conclusion. 



32 



Table 2-4. Interatomic Distances, Angles, and 71 Bond Dissociation Energies of 
Fluoroethenes. 





CH?— CH? 
1.339 


CH?=CHF 
1.333 


Cn2 — CF2 
1.316 


CHF=CF, 


CF,=CF, 


r(C=C),A 


1.309 


1.311 


r (C-F), A 


- 


1.348 


1.324 


1.336 


1.319 


ZHCH, deg. 


117.8 


114.7 


119.3 


- 


- 


ZHCF, deg. 


- 


111.3 


- 


114.0 


- 


ZFCF, deg. 


- 


- 


109.7 


109.1 


112.6 


71 D°, kcal mo!" 1 


63-64 


Unknown 


62.8 


Unknown 


52.3 



The marked decrease in FCF bond angles has been rationalized by Bernett 108 
and Kollman 109 on the basis of hybridization arguments; Epiotis has advanced an 
alternative explanation involving nonbonded attraction between fluorine atoms. 110,111 

\ F \ F + 

H — H 

Figure 2-2. Fluorine 2p Electron Derealization in Unsaturated Systems. 

Photoelectron 112 and electron attachment 113 spectral data for the fluoroethylene 
series are provided in Table 2-5. A significant lowering (over a range of 3.1 eV) of the a 
MO energies is observed, with only a slight (ca. 0.3 eV) variation in the u MOs. 
Stabilization of the a MOs is ascribed to extensive derealization over the fluorine 
substituents; such an interaction within the n system is diminished and counteracted by 
strong C-F antibonding overlap. 112 A steady increase in electron attachment energies 
with successive fluorination can also be seen, attributed to destabilization of n* resulting 
from an antibonding interaction with the fluorine 2p AOs. 113 

Heats of hydrogenation provided in Table 2-6 95 illustrate the reactivities of 
fluorinated alkenes. In general, transformation of a polyfluorinated olefin into a saturated 



33 



Table 2-5. Vertical Ionization Potentials (LP.) and Electron Attachment Energies (EA) for 
the Fluoroethenes. 



Ethene 


7il.P.,eV 
10.6 


al.P..eV 
12.85 


E.A.. eV 


CH 2 = Cn2 


1.78 


CH 2 =CHF 


10.58 


13.79 


1.91 


Cri2 = Cr2 


10.72 


14.79 


1.84 


c/s-CHF=CHF 


10.43 


13.97 


2.18 


trans-CHF=CHF 


10.38 


13.90 


2.39 


CF 2 =CHF 


10.53 


14.64 


2.45 


CF 2 =CF 2 


10.52 


15.95 


3.00 



derivative is more exothermic than for the parent hydrocarbon. This arises from a 
combination of the destabilizing effect of polyfluorination on double bonds and the 
thermodynamic preference for ge/n-difluoro substitution at saturated carbon. The 
deviation of CH 2 =CHF in Table 2-6 is explained by the preference of a single fluorine 
substituent to reside at the vinylic position (Figure 2-3). m 

Table 2-6. Heats of Hydrogenation of the Fluoroethenes. 

Ethene AH (H?). kcal mol" 1 
CH 2 =CH 2 -32.6 

CH 2 =CHF -29.7 

CH 2 =CF 2 -38.8 

CF 2 =CF 2 -45.7 

Fluorine Non-Bonded Interactions in Reactive Intermediates 

The n-donor ability of the fluorine substituent is reflected in its activating and 
ortho, para-directing character in electrophilic aromatic substitution reactions, 115 
consistent with 13 C NMR measurements of fluorobenzene, where shielding of these 
positions is observed. 116 






34 






H CH 2 F l 2 H 3 C F l 2 H 3 C H 

M - X - X 

H H AH° = -3.34 kcal mol" 1 H H AH = +0.65 kcal mol" 1 H F 

54 Z-55 E-55 



H 3 C F h H CF 2 H 

H F AH = -2.5 kcal mol -1 H H 

56 57 

Figure 2-3. Thermodynamic Equilibria in Mono- and Difluoropropenes. 

Derealization of fluorine's nonbonded electrons into the vacant 2p orbital on 
carbon more than compensates for its inductive withdrawal in a-fluoro carbocations, 
resulting in net stabilization. In the gas phase, carbocation stability increases along the 
series + CH 3 < + CF 3 < + CH 2 F < + CHF 2 and + CH 2 CH 3 « + CF 2 CH 3 = + CHFCH 3 . 117118 The 
+ CF 3 cation has been observed in the gas phase, with many others having been 
successfully generated in solution. 119 " 121 In constrast, fluorination at the p position and 
beyond destabilizes carbocations due to inductive effects; simple alkyl p-fluoro 
carbocations not benefiting from additional stabilizing factors have yet to be detected. 

Electron pair repulsion in a-fluoro carbanions results in a strong preference for 
pyramidal geometries, ao initio calculations 122 " 124 predicting an FCF angle of ca. 99.5° 
and an inversion barrier of 119 kcal mol" 1 for "CF 3 . Although fluorination does increase 
C-H bond acidities in such pyramidal systems, 125 a destabilizing effect is observed in 
cases such as the 9-fluorofluorenyl anion (Figure 2-4, X = F) where coplanarity is forced 
between the 2p orbitals on fluorine and the remainder of the n system. 126 Fluorine 
substitution in the p position stabilizes carbanions through a combination of inductive 
and hyperconjugative effects, (Figure 2-5) the latter supported by X-ray crystallographic 
data of perfluoroalkyl anion salts as well as through calculation. 122,127128 






35 




58 



NaOCH 3 
CH3OH 




59 



£i i fexc (reh 

D 1 

F 0.125 
CI 400 
Br 700 
Figure 2-4. Fluorine Destabilization in Planar 9-Halofluorenyl Anions. 



<Q 



c-c^i 

F 



>c=c< 



Figure 2-5. Negative Hyperconjugation in p-Fluorocarbanions. 



Fluorine Steric Effects 



The minimal spatial requirements of fluorine, the smallest non-hydrogenic 
substituent, would imply a very minor steric impact on reaction kinetics and 
thermochemistry. In most cases this is true; however, examples of steric inhibition in 
reactions of fluoro-substituted systems do exist, typically in conformational and other 
dynamic processes occuring via highly congested transition states. This is illustrated in 
Figure 2-6; 129130 the meta ring flip in 61 (X = F) exhibiting the largest known rate 
retardation induced by a single fluorine substituent. 

The disparate behavior observed in the Cope rearrangements of d,l- and meso- 
62 (Figure 2-7) 131 provides a particularly striking example of a fluorine steric effect. 
Transformation of d,/-62 to 63 proceeds via a typical chair-like transition structure, where 
in meso-62 a "boat-like" structure is required for d - C 6 bond formation. The higher AH* 



36 



X X 




X = H, AG* < 6 kcal mor 1 (340 K) 
X = F, AG* = 1 1 . 1 kcal mor 1 (340 K) 



60 




k H /k f =10 u (298 K) 



61 






Figure 2-6. Inhibition of Conformational Dynamics by Fluorine Substitution. 



and positive AS* implies a dissociative, rather than concerted, transition state for 
rearrangement of meso-62. This is induced by severe electrostatic repulsion between 
the high charge densities of the terminal fluorine substituents, separated by less than the 
sum of their van der Waals radii in the Cope transition state. 132 




d,/-62 




meso-62 



<?P 




AH* = 22.4 kcal n 
AS* = -17.5eu 



-1 




kj 



AH* = 49.5 kcal mol" 1 
AS* = +8.1 eu 





CF 2 

i 

CF 2 



63 





CF, 

i * 

CF, 



63 



Figure 2-7. Chair- versus Boat-Constrained Cope Rearrangement Reactions of 
Terminally Fluorinated Dienes d,l- and meso-62. 



37 

Steric effects are enhanced by perfluoroalkylation and branching. Cyclohexane 
A values 133 and modified Taft steric parameters 134 demonstrate that CF 3 is at least as 
large as isopropyl; evidence exists 135 to suggest that perfluoroisopropyl and re/t-butyl are 
comparable in size. The most remarkable example of the above affects is the existence 
of perfluorinated radical 64, (Figure 2-8) found to be persistent by ESR even in the 
presence of molecular oxygen. 136 The astounding kinetic stability of 64 derives from 
steric sheltering of the unpaired electron by the neighboring perfluoroalkyl groups. 




y F19 

64 

Figure 2-8. Scherer's Persistent Perfluoroalkyl Radical. 

The Fluorine Substituent in Free Radicals 

Early application of organofluorine radical chemistry was comprised of the chain- 
mediated addition of polyhalomethanes and ethanes to olefins, first by Haszeldine 137 and 
soon thereafter by Tarrant. 138139 Subsequent relative rate studies by Stefani et a/. 140 
followed by those of Tedder 141 142 clearly demonstrate the contrasting behavior of 
fluorinated and non-fluorinated radicals in their bimolecular additions to alkenes. 

The fluorine substituent has a substantial effect on the structure of organic 
radicals as well as dramatic, but comprehensible, alterations in hydrogen abstraction 
and addition reactivity in comparison to hydrocarbon systems, resulting primarily from 
fluorine's potent o-withdrawing character. 

Structural Aspects 

In contrast to the planar, n-type methyl radical, a-fluorination results in 
increasingly pyramidal, a-type radicals, as indicated by electron paramagnetic 



38 

resonance measurements 143 and ab initio theoretical studies. 144 " 147 Calculations by 
Pasto at the UHF/4-31G level indicate barriers to inversion of 0.5, 6.8, and 25.1 
kcal mol" 1 for *CH 2 F, 'CHF 2 , and *CF 3 , respectively. 147 Geometries of the fluoromethyl 
radicals computed at the UHF/6-31G(d) level are provided in Figure 2-9. 



I 90.0° '101.1° j 105.8° I 107.6° 

II I | 

■^ '^ '^ l"> 







Figure 2-9. UHF/6-31G(d) Pyramidalization Angles of *CH 3 , *CH 2 F, *CHF 2 , and *CF 3 . 

Inversion barriers in the series appear somewhat sensitive to the level of theory 
employed and substantially increase with the inclusion of polarization functions. SCF 
calculations by Dykstra 145 employing a polarized double-^ basis result in an inversion 
barrier of 33 kcal mol" 1 for *CF 3 ; inclusion of electron correlation (QCISD(T)/ 
6-31G(d)//UHF/6-31G(d), present work) affords a value of 29.4 kcal mol" 1 . Such 
successive deviation from planarity is due to a combination of effects; relief of In 
repulsion between the singly-occupied orbital on carbon and the fluorine 2p electrons is 
further reinforced by overlap between the carbon 2p and C-F antibonding orbitals. This 
strong tendency for pyramidalization has been shown to be responsible for the low n 
bond energy in tetrafluoroethylene (Tables 2-4 and 2-6). 148 

The structures of fluorinated C 2 radicals have been theoretically probed by 
Paddon-Row and Chen er a/. 149 " 152 Alkyl substitution induces slight pyramidalization as 
observed in the ethyl radical, (Figure 2-10) due to hyperconjugation between the SOMO 
and the staggered p C-H bond. 153 Fluorination at the radical center exerts an effect 
similar to that observed in the methyl series, while the structures of alkyl radicals are 
found to be relatively insensitive to p-fluorination. 









39 





I 94.40 





/ 105.9° 



Figure 2-10. UHF/6-31G(d) Pyramidalization Angles for Ethyl Radicals CH 3 CH 2 *, 
CF 3 CH 2 *, CH 3 CF 2 # , and CF 3 CF 2 *. 



Radical Stability 

FMO theory dictates that for radicals bearing electronegative substituents with 
lone pairs (F, OH, NH 2 , SH) an inductive, destabilizing influence exists, countered by 
stabilization resulting from derealization of the unpaired electron. 154 Thus, in the a 
sense, fluoroalkyl radicals are destabilized. Furthermore, the opposing ^-stabilizing 
effect of the fluorine lone pairs decreases with pyramidalization of the radical site, due to 
diminished overlap with the 2p AO on carbon. 

The progressive decrease in stability of alkyl radicals with a- or p-fluorination has 
been illustrated by Pasto, 80147 in the form of calculated radical stabilization energies 
(RSE) based on isodesmic reactions (Table 2-7). The aforementioned increase in C-H 
bond dissociation energies along the fluoromethane series lends experimental support. 

Although some degree of p C-F hyperconjugative interaction is observed in 2- 
fluoro substituted radicals, 152 such a contribution to overall radical stability is minor and 
inductive destabilizing influences dominate, as indicated by experimental and theoretical 
C-H BDEs for the 2-fluoroethanes (Table 2-8). 



40 



Table 2-7. Isodesmic Equation and Radical Stabilization Energies (RSE, kcal mol" 1 , 
4-31 G) for a- and p-Substituted Systems. Positive Values Denote Radical Stabilization. 

(2-1) 



X n CH3. n + 


CH 4 - 


-> X n CH 4 . n + 


*CH 3 


X RSE 




X 


RSE 


F +1.64 




CH 3 


+3.27 


F 2 +0.56 




OH 


+5.73 


F 3 -4.21 




OCH 3 


+5.30 


CH 2 F +1.46 




CN 


+5.34 


CHF 2 +0.16 




NH 2 


+10.26 


CF 3 -1.34 




+ NH 3 


-4.07 






SH 


+5.66 






+ SH 2 


-3.17 



Table 2-8. Experimental and Theoretical C-H Bond Dissociation Energies (kcal mol" 1 ) 
of 2-Fluoroethanes. 



n3UH 2 -H 


L>H 2 rL>H 2 - 


■H 


CF?HCH?- 


H 


CF 3 OH 2 -H 


101.1** 


103.6' 








106.7 af 


97.7 C 


99.6 C 




101. 3 C 




102.0 C 


100.0 d 










104.3 d 


102.0 e 


104.1 e 




105.9 e 




107. 1 e 



3 Experimental Value; Reference 155. b Experimental Value, Reference 156. 
c MP2/6-311G(d,p)//MP2/6-31G(d,p); Reference 157. d B3LYP/6-31G(d); Reference 91 
and Present Study. e MP2/6-311+G(3df,2p)//MP2/6-31G(d); Reference 158. 
'Experimental Value; Reference 159. 



Organofluorine Radical Reactivity 

As a result of the a-withdrawing character of the fluorine substituent and 
interaction of the SOMO with C-F a* orbitals, fluoroalkyl radicals should possess lowered 
SOMO energies and therefore exhibit enhanced SOMO-HOMO interactions in 
comparison with reactions of their hydrocarbon counterparts. Experimental ionization 









41 



potential and electron affinity data, although sparse, has been compiled in a recent 
review by Dolbier 90 and demonstrates the greater absolute electronegativity of the 
fluoroalkyl radicals. Calculated quantities, inferred from Koopmans' theorem 160 or based 
on radical-ion energy differences, follow the expected trend although quantitative 
agreement is often lacking. 161162 The combination of such FMO, geometric, and 
enthalpy factors in hydrogen atom abstraction, intermolecular addition, and cyclization 
reactions of fluoroalkyl radicals is now discussed. 

Hydrogen Atom Abstraction Reactions 

A review by Tedder 163 has underlined the importance of polar and enthalpic 
factors in radical abstraction reactions. Activation parameters for abstraction by methyl 
and trifluoromethyl radicals from a series of hydrogen donors are provided in Table 2-9. 



Table 2-9. Arrhenius Parameters for Hydrogen Atom Abstraction by Methyl and 
Trifluoromethyl Radicals. 







CH 3 * 




CF 3 * 


H-Donor 




E a 


loq A 




E a a loq A 


CH 3 -H 




14.2 


8.8 




11.3 8.9 


CH3CH2-H 




11.8 


8.8 




6.9 8.4 


(CH 3 ) 2 CH-H 




10.1 


8.8 




6.5 8.1 


(CH 3 ) 3 C-H 




8.0 


8.3 




4.9 7.7 


H-CI 




2.5 b 






5.0 C 


In kcal mol" 1 . 6 AH° rx „ = 


-1 kcal 


mol" 1 . 


AH pen = 


-3 kcal 


mol" 1 . 



In the first four examples, the decrease in activation barrier for both CH 3 " and 
CF 3 * abstractions from alkanes are in line with the greater stability of the product radical; 
in each case, the barrier to abstraction by CF 3 * is substantially lowered. In contrast, 



42 

abstraction of hydrogen atom from HCI by CF 3 * is much less facile, occuring with a 
barrier twice that of CH 3 * in spite of its slightly greater exothermicity. 

In collaboration with Lusztyk and Ingold at NRCC, LFP-determined rates of 
perfluoroalkyl radical addition to a number of alkenes have been determined by the 
Dolbier group. 54,858891 This has afforded, via competitive kinetic techniques, absolute 
rate constants for hydrogen abstraction by the perfluoro-n-heptyl radical from a series of 
donors, summarized in Table 2-1 0. 54868789 For comparison, abstraction rate constants 
for hydrocarbon n-alkyl radicals were provided in Table 1-1. 



Table 2-10. Absolute Rate Constants for Hydrogen Atom Abstraction for Perfluoro-n- 
heptyl Radicals. Rate Constants are at 303 K. 

Et 3 SjH (TMS) ? Si(CH ,)H nBu 3 GeH (TMS^SiH nBu ^ SnH PhSH 

/c H (n-C 7 F 15 *), 0.75 16 15 51 203 0.28 

10 6 M" 1 s- 1 



For the first five donors in the series, a substantial rate enhancement is observed 
over hydrocarbon radicals, ranging from 75-fold in the case of nBu 3 SnH to nearly 900- 
fold in the case of Et 3 SiH, after slight temperature correction to 303 K. Although 
hydrogen transfer to perfluoroalkyl radicals is more exothermic (see the previous 
discussion on C-H BDEs) this is insufficent to account for a nearly three order of 
magnitude difference in rate constants. Furthermore, thiophenol, an excellent donor to 
hydrocarbon radicals, is found to suffer a greater than 400-fold decrease in transfer rate 
to perfluoroalkyls. 

These characteristics are explained by the ability of the radical-donor pair to 
accommodate charge transfer interactions in the hydrogen transfer transition state, as 
previously discussed in Chapter 1. The relatively electropositive donor agents 
(stannanes, germanes, and silanes) lead to a more favorable polarity matchup with the 
electronegative perfluoroalkyl radical than with the hydrocarbon. Conversely, transfer 
from the more electronegative thiophenol results in a non-polarized or polarity- 



43 



mismatched transition state. The existence of such polar effects were confirmed by a 
correlation between rates of hydrogen transfer to perfluoro-n-heptyl radicals by a series 
of substituted thiophenols, versus their Hammett a+ constants (Figure 2-1 1). 87 The 
resulting p value of -0.56, when compared to that obtained in the case of terf-butoxyl 
(-0.30) 164 again reflects the high electrophilicity of perfluoroalkyl radicals. 




Figure 2-11. Hammett Plot for Hydrogen Abstraction from para-Substituted Thiophenols 
by Perfluoro-n-heptyl Radical. 



Intermolecular Addition Reactions 

Relative rates of addition of small fluorocarbon radicals to fluorinated and non- 
fluorinated olefins have been extensively investigated by Tedder and Walton, 
culminating in a critial review in 1980. 52 Table 2-1 1 illustrates the relative reactivity of 



Table 2-11. Relative Rates of Addition of the Fluoromethyl Radicals to Ethene and 
Tetrafluoroethene. 



Radical 


kartd (C7F4 


lk^(CM.) (437 K) 


*CH 3 




9.5 


'CH 2 F 




3.4 


'CHF 2 




1.1 


*CF 3 




0.1 



44 

the fluoromethyl radicals towards ethylene and tetrafluoroethylene. Additional studies, 
utilizing a number of unsymmetrical methyl- and trifluoromethyl-substituted olefins, 
solidified the ascription of relative rates and regiochemical preferences to a combination 
of polar and steric influences. 

Absolute rates of addition of perfluoroalkyl radicals to alkenes have been 
determined by laser flash photolysis, a subset of the data acquired to date presented in 
Table 2-1 2. 90 The dramatic rate acceleration enjoyed by the perfluoroalkyl radicals 
versus hydrocarbon n-alkyls is readily apparent, ranging from factors of 300 to 30,000 in 
the case of the heptafluoropropyl radical addition to electron-rich alkenes. The rate of 
addition to pentafluorostyrene, in constrast, is increased by only a factor of 42. 

Table 2-12. Absolute Rate Constants for Addition of Perfluoroalkyl Radicals to Alkenes. 

k ari H , 10 6 M 1 s 1 (298 K) 
Alkene X 3 F Z X7F15 ICgF^ X2F5 XF3 RCH/ 

Styrene 43 46 46 79 53 0.12 

a-Methylstyrene 78 89 94 87 0.059 

P-Methylstyrene 3.8 3.7 7.0 17 

Pentafluorostyrene 13 23 26 0.31 

4-Methylstyrene 61 

4-Methoxystyrene 65 

4-Chlorostyrene 36 

4-(CF 3 )Styrene 35 

1-Hexene 6.2 7.9 16 0.0002 

Such enhancements may potentially be attributed to a combination of factors. 
Relative reaction enthalpies should play a role, as a stronger C-C bond (from CH 3 -CH 3 
versus CF 3 -CH 3 BDE data in Table 2-3, ca. 1 1 kcal mol" 1 ) is formed upon perfluoroalkyl 



45 

addition. However, only slight increases (factors of 5-7) in addition rates of the 
perfluoroalkyls to styrene versus 1-hexene are observed, despite the greater 
exothermicity (ca. 16 kcal mol" 1 ) of the former. This demonstrates the relatively minor 
importance of reaction enthalpy, in accord with the early transition states expected for 
radical addition. 

The pyramidal, a-character of the fluoroalkyl radicals should afford a kinetic 
advantage (further discussed in Chapter 3) over the planar hydrocarbon, the LFP- 
measured rate of addition of 1,1-difluoropentyl radical to styrene, 2.7 x 10 6 M" 1 s" 1 , 88 
giving rise to a 22-fold enhancement relative to n-alkyls. 

The primary factor responsible for such striking increases in reactivity is believed 
to be charge transfer influences (Figure 2-12) similar to those postulated for hydrogen 
atom transfer reactions. The lowered SOMO energies of the perfluoroalkyl radicals and 
resulting enhanced SOMO-HOMO interactions with alkenes in the addition transition 
state leads to substantial rate enhancement. Supporting evidence, in the form of a 
Hammett relation involving para-substituted styrenes and correlations between addition 
rates and alkene ionization potentials, has been offered. 54 Ab initio computations concur 
with the experimental findings and are discussed in Chapter 3. 

5- 

CF 3 (CF 2 ) n CF2 

8+ 



Figure 2-12. Polarized Transition State for Addition of Perfluoroalkyl Radicals to 
Alkenes. 



Intramolecular Cyclization Reactions 

Cyclopolymerization of fluorinated monomers has long been known as a means 
for the generation of macromolecular materials with unique physical properties. 165 " 168 



46 

However, despite the widespread popularity of the unimolecular 5-hexenyl radical 
cyclization for the generation of five-mem bered rings, synthesis of fluorinated cyclic 
products utilizing radical methodology has received limited attention. 90 This is somewhat 
surprising, in light of the current interest and demonstrated importance of fluorinated 
analogues and mimics of pharmaceutical and agricultural agents. 169 

Until recently, no quantitative kinetic data existed for cyclization reactions of 
fluorinated radicals. With the competitively-determined rate constants of hydrogen 
abstraction by perfluoroalkyl radicals serving as basis reactions, rates of cyclization of a 
number of fluorinated 5-hexenyl systems have now been determined, 86 ' 89,170 examples of 
which are provided in Figure 2-13. 

Most obvious of the data is the remarkable rate acceleration in the 5-exo 
cyclizations of 65 and 68, occurring with 163- and 41 -fold increases relative to the parent 
hydrocarbon 1 and ascribed to charge-transfer effects analogous to those occurring in 
bimolecular additions. Consequently, with no such polarity matchup in 71 (which 
involves cyclization of an electrophilic radical onto an electron-deficient double bond) 
only a minor increase in rate is observed. The 5-exo cyclization rates of 71 and 74, in 
line with those other hexenyl systems bearing fluorinated double bonds, 170 demonstrate 
the lack of kinetic impact of vinylic substitution. 

Especially surprising is the degree to which 65 and 68 undergo 6-endo closure, 
the former with a 1040-fold, and the latter a 700-fold acceleration relative to 1. This is 
further discussed in Chapter 4, in which the reactivities of a series of lightly-fluorinated 5- 
hexenyl systems are investigated. 

In the preceeding discussions of hydrogen abstraction, addition, and cyclization 
reactivity, the observed kinetic behavior was found to be due to a combination of 
geometric and polar effects induced by polyfluorination. Related studies of partially 
fluorinated alkyl radicals, addressed in the next two chapters, will aid in separating the 
effects of fluorination at the a and p positions and beyond, providing insight into the 



47 

extent to which the effects of perfluorination on the reactivity of organic radicals are the 
sum of their parts. 



Table 2-13. Some Absolute Rate Constants for 5-exo and 6-endo Cyclization of 
Fluorinated 5-Hexenyl Radicals. Rate Constants are for 303 K. 

Cyclization Reaction k C5 , 10 5 s" 1 k C6 , 10 5 s" 1 





F CF 2 * 
F 2 C C .CF 2 

F 2 
71 

F 

F . 




F T 

F 2 C.J 



2 ^C 
F 2 




2.7 0.05 



CF 2 
F 2 F 2 C-CF 2 '^""C 



Q^ * .(? + F 2 C^CF 2 44 ° 52 



F 2 
65 66 67 



^ cf 2 a r C F 2 

F 2 C. C .CF 2 ^ FC-CF F 2 C ^ CF 2 HO 35 

F 2 2 2 F 2 

68 69 7 o 

F 



F x cp; 

F 2 C CF 2 
F 2 C-CF 2 


F 
l. 
F 2 C CF, 

4- J ' 

F 2 C. C „CF 2 
F 2 


72 


73 




F 


F y CF 2 ' 
F 2 C— ' 


F 2 C CF 2 



4.9 N/A 



4.3 N/A 



74 75 76 



48 
Conclusion 

An introduction to the general structural and reactivity characteristics imparted by 
fluorine substitution in organic systems has been provided. Such substitution can either 
be stabilizing or destabilizing, depending on the nature of the ground state molecule, 
intermediate, or reaction in question. 

A summary of the first absolute kinetic data obtained for reactions of fluorinated 
radicals has been presented. Results of these studies, involving per- or otherwise highly 
fluorinated systems, demonstrate the combined influence of polarity, structural, and 
enthalpic factors. The following studies of the addition, hydrogen abstraction, and 
cyclization kinetics of partially fluorinated alkyl radicals will serve to dissect the relative 
magnitudes of these influences on organic radical reactivity. 









CHAPTER 3 

THE REACTIVITY OF PARTIALLY FLUORINATED RADICALS 
IN INTERMOLECULAR ADDITION AND HYDROGEN ABSTRACTION REACTIONS 



Introduction 

Initial studies of the addition rates of some partially-fluorinated radicals to 
alkenes 88 demonstrated observable rate enhancements relative to n-alkyls, though not 
nearly as great as those of perfluoroalkyl systems. Such reactivity derives from the 
combination of structural and polar characteristics induced by fluorine substitution. 

The current study extends the amount of absolute rate data assembled for the 
addition of partially fluorinated radicals to olefins. In addition, through competitive kinetic 
techniques, absolute rate constants for hydrogen abstraction from tri-n-butyltin hydride 
(nBu 3 SnH) have been determined. Such kinetic data is necessary for the determination 
of absolute rates of cyclization of partially fluorinated 5-hexenyl radicals, discussed in 
Chapter 4, and allows for the partitioning of the gross reactivity characteristics of 
perfluoralkyl radicals into the separate influences of a, p, and y fluorination. This is 
accomplished by use of the following systems: RCH 2 CH 2 CF 2 * (a,a-difluoro), 
RCH 2 CF 2 CH 2 * (p,p-difluoro), RCH 2 CF 2 CF 2 * (a,a,p,p-tetrafluoro), CF 3 CF 2 * (a.a.p.p.p- 
pentafluoro), and RfCH 2 CH 2 " (y-perfluoro). 

The existence of charge transfer stabilization in the transition states for 
fluorinated radical addition to alkenes is corroborated by ab initio calculations, and the 
thermodynamics of C-H and C-C bonding and radical stablization in hydrofluorocarbons 
rationalized on the basis of Coulombic interactions. 



49 



50 
Precursor Syntheses and Competitive K inetic Studies 

In each of the competitive kinetic runs, hydrofluorocarbon radicals were 
generated from bromide or iodide precursors by photoassisted C-X bond homolysis. 
These radicals subsequently underwent competitive trapping with known, varying 
concentrations of styrene or nBu 3 SnH, adjusted to ensure pseudo-first-order kinetic 
behavior and to allow for accurately measurable amounts of trapping products, as 
depicted in Figure 1-17 and in greater detail below. 

1.1-Difluorohex-1-vl Radical (77) 

The synthesis of bromide precursor 80 was achieved in two steps (Figure 3-1) in 
a straightforward manner. Copper(l)-mediated addition of dibromodifluoromethane to 1- 
pentene (78), based on a modification by Gonzalez et a/. 171 of a procedure by Burton 
and Kehoe 172 afforded 1,3-dibromo-1,1-difluorohexane 79 in typical yield. 
Regioselective displacement of the internal bromine was accomplished via sodium 
borohydride reduction in DMSO, providing precursor 80 contaminated with a small 
amount of overreduction product 81. Pure samples of each were obtained by 
preparative GC separation, the former utilized in the competition run and the latter for 
spectral comparison with kinetic NMR data. 



CF 2 Br 2 




(CH 3 ) 3 COH, H 2 NCH 2 CH 2 OH 

CuCI(cat.) (57.3%) 

78 79 



NaBH 4 




CF 2 H 



DMSO 

80 81 

Figure 3-1 . Preparation of 1-Bromo-1 ,1-difluorohexane 80 and 1 ,1-Difluorohexane 81 



51 



,CF 2 Br 



80 



hv 

nBu 3 Sn(H 
C 6 D 6 



77 



nBu 3 SnH 



C5H5CH— CH2 



^add 



CF 2 H + nBu-^Sn" 



81 



.& 



82 




nBu 3 SnH 



F r 2 

^^"if^l + nBu 3 Sn ' 

83 y* 

Figure 3-2. k H I k aM Competitive Kinetic Scheme for 1,1-Difluorohex-1-yl Radical 77. 

Photolysis of 80 as a C 6 D 6 solution in the presence of an excess of nBu 3 SnH and 
styrene (Figure 3-2) afforded intermediate radical 77. Subsequent entrapment by these 
agents (both irreversible processes) yielded 81 and 82, respectively; the latter further 
trapped by nBu 3 SnH to yield 3,3-difluoro-1-phenyloctane 83. Throughout the course of 
the reaction, nBu 3 Sn* radicals are generated to propagate the chain process via 
abstraction of halogen from precursor 80. 

Product ratios for varied concentrations of trapping agents were determined by 
19 F NMR analysis according to the pseudo-first-order relation in Equation 3-1, 



[81] 
[83] 



[ *H ] [ 77 ] [ nBu 3 SnH ] 
[ *add ] [ 77 ] [ C 6 H 5 CH=CH 2 ] 



(3-1) 



a plot of product ratios obtained for each data point versus that of trapping agents 
affording the ratio k H I k aM . The stability of trapped products under the reaction 
conditions and lack of appreciable side reactions were demonstrated by the high 



52 

conversion of precursor 80 to 81 and 83 versus an internal standard of a,ot,a,- 
trifluorotoluene, (<j> -63.24) indicating in turn the high efficiency of the chain process and 
reliability of the kinetic results. A partial 19 F NMR spectrum of the first of six data points 
is provided in Figure 3-3, a doublet of triplets (-CF 2 H, $ -116.0) observed for 81 versus 
an overlapping triplet of triplets at f -99.1 (-CF 2 -) for 83. Full kinetic data and yields are 
given in Table 3-1 below, the plot of which located in Figure 3-4. The slope of the line 
(3.39 ± 0.02) in conjunction with the known absolute rate constant for addition of 






-98 -99 -100 -101 -102 -103 -104 -105 -106 -107 -108 -109 -110 -111 -112 -113 -114 -115 -116 -117 -118 

Figure 3-3. Partial 19 F NMR Spectrum of Data Point 1 for k H I k aM Competition of 1,1- 
Difluorohex-1-yl Radical 77. 



Table 3-1. Competitive Kinetic Data for k H I k aM Competition of 1,1-Difluorohex-1-yl 
Radical 77. 

r 80 1 f C fi H s CH=CH 7 1 f nBu . SnH 1 / fC fi H s CH=CH ? 1 f 81 1 / [ 83 1 % Yield 



0.094 


2.01 


0.719 


2.30 


95 


0.094 


1.81 


0.847 


2.73 


95 


0.094 


1.61 


1.01 


3.26 


96 


0.094 


1.41 


1.21 


3.93 


97 


0.094 


1.21 


1.49 


4.88 


96 


0.094 


1.01 


1.87 


6.21 


95 



53 

1 , 1 -difluoropent-1 -yl radical to styrene, 2.7 (± 0.5) x 10 6 M" 1 s'\ resulted in an absolute 
rate constant k H of 9.1 (± 1.7) x 10 6 M" 1 s' 1 . It should be noted that the accuracy of such 
derived k H values can be no better than those reported in the LFP determinations of fr add , 
the error estimates in the former reflecting both the least-squares fit of the line and 
propagated error of the latter. Synthesis of styrene adduct 83 was performed by classic 
means for characterization and spectral comparison (Figure 3-5). 



CO 
00 



00 




0.50 



0.75 



1.00 1.25 1.50 

[ nBu 3 SnH ] / [ C 6 H 5 CH=CH 2 ] 
Figure 3-4. Plot of the Data in Columns 3 and 4 of Table 3-1 . 



1.75 



2.00 




Br 



Mg 



Et,0 



84 




MgBr 1. CH 3 (CH 2 ) 4 CHO 

2. H 3 + 



85 




Na 2 Cr 2 7 / H 2 S0 4 
Et,0 




DAST 



CH 2 CI 2 



(68.0%) ^^ 
83 



Figure 3-5. Preparation of Styrene Adduct 83. 



54 
2,2-Difluorohex-1-yl Radical (88) 

In accordance with literature precedent, 173 a-bromination of 2-hexanone in the 
presence of urea in acetic acid selectively afforded 1-bromo isomer 90 in 72.2% yield. 
Subsequent treatment with diethylaminosulfurtrifluoride (DAST) provided bromo 
precursor 91, originally purified by preparative GC for use in the kinetic study. However, 
a sluggish chain reaction (further hindered by the strong UV absorption of the excess 
styrene present in the kinetic samples) led to the preparation of iodo precursor 92 
(Figure 3-6) via Finkelstein transformation at elevated temperature. 



o 



89 



Br 2 



CH 3 C0 2 H, H 2 NCONH 2 



O 



s CH 2 Br 



(72.2%) 
90 



DAST 



CHCI-, 



C 

F 2 



CH 2 Br 



Nal 



(CH 3 ) 2 CO 




CHol 



(59.7%) 
91 



Figure 3-6. Preparation of 1-lodo-2,2-Difluorohexane, (92) Precursor to 2,2-Difluorohex- 
1-yl Radical 88. 




0.20 



0.22 



0.32 



0.24 0.26 0.28 0.30 

[ nBu 3 SnH ] / [ CgH 5 CH=CH 2 ] 

Figure 3-7. Plot for k H I k aM Competition of 2,2-Difluorohex-1-yl Radical 88. 



0.34 



55 

The competition plot for k H I k aM determination is found in Figure 3-7; raw data for 
this and all remaining experiments in this chapter may be found in Chapter 5. With the 
absolute rate constant for addition of 2,2-difluoropent-1-yl radical to styrene known from 
LFP experiments, a k H value of 1.4 (± 0.5) x 10 7 M" 1 s" 1 was determined. Authentic 
samples of 93 and 98 were prepared for spectral comparison and characterization as 
illustrated below in Figure 3-8. 



F 2 
91 



CH 2 Br 



nBu 3 SnH 



AIBN 
CeHg 



,CH, 



93 




Br 



94 



Mg 



EtoO 





MgBr 



95 



Na 2 Cr 2 7 / H 2 S0 4 
Et?0 



(71.1%) 




96 




DAST y^ 

OH2CI2 


F 2 




(66.3%) 




98 



^ 




1. CH 3 (CH 2 ) 3 CHO 

2. H 3 + 




O 



(83.7%) 
97 



Figure 3-8. Preparation of Hydrogen Abstraction Product 93 and Styrene Adduct 98. 



1.1.2.2-Tetrafluorobut-1-vl (99) and 1.1.2.2-Tetrafluorohex-1-yl (100) Radicals 



At the time of this study, no absolute rate constants for addition of a 1,1,2,2- 
tetrafluorinated radical to alkenes had been determined. Thus, in order to obtain a k H 
value for such a system, a precursor suitable for absolute k aM measurements was first 
required. Bromide precursors, although in most cases sufficient for competition 



56 

experiments, are ineffective under the LFP operating conditions used in our k aM 
determinations (308 nm excimer laser pulses) due to their relatively short wavelength 
chromophore and small extinction coefficient (for 102, e max = 37.3 M" cm" , X max = 218 
nm, cyclohexane solvent). The instability of O-acylthiohydroxamic esters of the 
perfluoroalkanoic acids has been noted by Barton. 174 With neither these nor diacyl 
peroxide precursors lending themselves to isolation and / or shipment to the NRCC in 
Canada, the synthesis of a suitable iodide precursor was undertaken. 

1-Bromo-1,1,2,2-tetrafluorohexane (102, Figure 3-9) was prepared in one step 
from 6-bromo-5,5,6,6-tetrafluorohex-1-ene (supplied by Halocarbons, Inc.) and its 
transformation to the corresponding iodide or carboxylic acid (the latter of which could be 
converted to the iodide via Hunsdieker methodology) attempted under a variety of 
conditions (Figure 3-10). 

1. BH 3 Me 2 S 

2. CsHnCOzH 





Tetraglyme 

(79.6%) 

101 102 

Figure 3-9. Preparation of 1-Bromo-1,1,2,2-tetrafluorohexane (102). 

Although the conversion of perfluoroalkyl iodides to their lighter analogues is 
known in the literature, 175 ' 176 downward transhalogenation of perfluoroaliphatic halides is 
exceedingly difficult. Indeed, all attempts at conversion of 102 to the corresponding 
iodide were met with failure. Perfluoroalkyl Grignard reagents, generated either directly 
(and in low yield) or via transmetallation by an alkylmagnesium halide, utilize iodide 
starting materials. 177 178 Investigations of perfluoroalkylzinc halides by Miller 179 resulted 
in a similar conclusion; perfluoro-n-propyl iodide may be converted to the organozinc 
reagent in ca. 60-80% yield (as determined by aqueous hydrolysis or capture with 
halogen electrophiles) after a brief induction period. On the contrary, reaction of 
heptafluoro-n-propyl bromide with zinc in 1,2-dimethoxyethane afforded no product after 



102 



CF 2 Br 



F 2 
102 



CF 2 Br 



78 




103 



57 



1) Mg 






2) l 2 


N. R. 




Et 2 






1) f-BuLi, -80°C 

2) l 2 




r 


Et 2 


^ 




1) f-BuLi, -100°C 




j 


2) l 2 




// 


Et 2 
1) MeLi, -100°C 


^ 


~ V\ 




V\^_ 


2) l 2 




\ 


Et 2 


^ 


\ 


1) MeLi, -100°C 




\_ 


2) C0 2 


— *> 



Et,0 



ICF 2 CF 2 I 




tBuOH, H 2 NCH 2 CH 2 OH 
CuCI (cat.) 



N. R. 



Et 3 N 3HF/NIS 
CH 2 CI 2 , 0° C 



N. R. 






Figure 3-10. Attempted Preparation of lodo Analogue of 102. 



103 



73 hours at 90° C; a 60% yield of the zinc derivative was obtained (inferred via 
hydrolysis) after a period of 1.5 months. Attempted lithiation of 102 at low temperature 
resulted in the formation of p-fluoride elimination product 1,1,2-trifluoro-1-hexene, (103) 
as identified by its 19 F NMR spectrum. Additionally, neither Cu(l)-induced addition of 
1 ,2-diiodotetrafluoroethane to 1-pentene nor iodofluorination of 103 utilizing the 
triethylamine trihydrofluoride / N-halosuccinimide methodology of Alvernhe et a/. 180 were 
successful. 



58 

Synthesis of a C 4 iodide was achieved through modification of a DuPont literature 
procedure, 181 whereby 1,4-diiodo-1,1,2,2-tetrafluorobutane was produced in 40.1% yield 
via direct thermal addition of 1 ,2-diiodotetrafluoroethane to ethylene (Figure 3-11). DBU- 
induced elimination of hydrogen iodide in ether followed by diimide hydrogenation with 
hydrazine-hydrogen peroxide in methanol afforded tetrafluoroiodo precursor 107 in 2.8% 
overall yield after preparative GC purification, which was sent to the NRCC for absolute 
k 3M measurements. 



ICF 2 CF 2 I 
104 



H2C— CH2 



I. 



CF, 



' C 
F 2 

(40.1%) 
105 



DBU 



Et 2 



F 2 



CFol 



(66.0%) 
106 



HN=NH 



CH3OH 



F 2 



CF 7 I 



(10.7%) 
107 






Figure 3-11. Preparation of 1,1,2,2-Tetrafluoro-1-iodobutane (107). 




, , 


6 - 


Coefficients: 







m = 4.56 


^ 






*■ 


5 - 


b =0.115 


- ' 




r 2 = 0.998 


, . 






00 


4 - 











*p 


3 - 





[ nBu 3 SnH ] / [ CgH 5 CH=CH 2 ] 
Figure 3-12. Plot for k H I k aM Competition of 1,1,2, 2-Tetrafluorohex-1-yl Radical 100. 



Bromide 102 was utilized in the k H I k aM competition, the kinetic plot for which 
given in Figure 3-12. An authentic sample of styrene addition product 110 was prepared 



59 

by slow syringe pump addition of nBu 3 SnH to a heated, irradiated solution of 102 and 
styrene in benzene (Figure 3-13). 



F 2 
102 



CF 2 Br 



nBu 3 SnH 
CeH6 



C5H5CH— CH2 

nBu 3 SnH 
CeHe 



nBu 3 SnH 



F 2 
108 



CF,H 



F 2 



i 2 




109 



C 
F 2 



& 




^ 



110 



Figure 3-13. Preparation of Hydrogen Abstraction Product 108 and Styrene Adduct 110. 

2-fPerfluorohexvlleth-1-vl Radical (111) 

2-[Perfluorohexyl]-1-iodoethane 112 was provided as a gift from Prof. Neil Brace. 
The k H I k aM competition plot is found in Figure 3-14; hydrogen abstraction product 
1 -[perfluorohexyl]ethane (113) and styrene adduct 1-[perfluorohexyl]-4-phenylbutane 
(117) were prepared as shown in Figure 3-15. 





6 - 


Coefficients: -— ♦""'^ 


r<- 




m=16.0 ^^^^^ 


£ 


S - 


b =-0.371 ^^^^ 


■— • 




r 2 = 0.999 ^^-*^ 


CO 


4 - 




^ 






^ 








3 - 







c 


I I I I 



0.20 



0.45 



0.25 0.30 0.35 0.40 

[ nBu 3 SnH ] / [ CgH 5 CH=CH 2 ] 

Figure 3-14. Plot for k H I k aM Competition of 2-[Perfluorohexyl]eth-1 -yl Radical 111. 



CgF-^Cr^Cry 
112 

C 6 F 13 I + 
114 

NaBH 4 



nBu 3 SnH 
A 




115 



60 

C6F13CH2CH3 
113 

Et 3 B (cat.) 




C fi F 



6 r 13 




(82.8%) 
116 



DMSO 

(70.0%) 

117 

Figure 3-15. Preparation of Hydrogen Atom Abstraction Product 113 and Styrene 
Adduct117. 



Pentafluoroethyl Radical (118) 

lodopentafluoroethane (119) was obtained from PCR, Inc. Due to the high 
volatility of both this precursor (bp 12-13° C) and hydrogen atom abstraction product 
pentafluoroethane (120, bp -48.5° C) 119 was handled as a solution in degassed C 6 D 6 . 
The k H I k aM competition experiment (Figure 3-16) was performed in tubes which were 
quickly flame-sealed upon injection of an aliquot of the chilled precursor stock solution. 



CM 
CM 



O 
CM 



2.0 - 


Coefficients: ^ — -"^"^ 


1.8 - 


m = 2.62 ^^-^^' 
b = 0.027 ^*~ — "• 


1.6 - 


r'= 0.998 ^^j*^^ 


1.4 - 


^^^*^^ 


1.2 - 


^^^* 


1 


"^ 




I I I I I I I 



0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 
[ nBu 3 SnH ] / [ CgH 5 CH=CH 2 ] 
Figure 3-16. Plot for k„ I k aM Competition of Pentafluoroethyl Radical 118. 



61 

Samples of 120 and 122 were prepared under free radical conditions for characterization 
purposes as shown in Figure 3-17. 



hv 



CF 3 CF 2 I 
119 



nBu 3 SnH 



C5H5CH— CH2 

nBu 3 SnH 
CgH6 



CF 3 CF 2 H 
120 



F ' c ' c ^O 

121 



nBu 3 SnH 



F 3 C 



£ 




122 



Figure 3-17. Preparation of Hydrogen Atom Abstraction Product 120 and Styrene 
Adduct 122. 



Discussion 



Absolute rate constants for addition and hydrogen atom abstraction for systems 
studied in this chapter are provided in Table 3-2. For comparison, data for hydrocarbon 
(n-pentyl) and perfluoroalkyl (perfluoro-n-heptyl) radicals are included. 

It is seen from the data that the reactivity trends for partially fluorinated radical 
additions to styrene are generally adhered to in hydrogen abstractions. However, the 
actual rate constants for the latter are observed to differ (on the average, by a factor of 
11) from those for addition, and span a narrower range. The decrease in abstraction 
rate ratios as a function of substitution is due to the proximity (within an order of 
magnitude) to diffusion control for the more reactive radicals 100, 118, and 127. 

C-H and C-C bond dissociation energies of the fluoroalkanes should reflect the 
relative thermochemistry of hydrogen atom abstraction and addition to alkenes by their 
respective radicals. However, very few experimentally determined BDE values for such 



62 



Table 3-2. Absolute Rate Constants for Hydrogen Abstraction from Tributyltin Hydride 
and Addition to Styrene by Partially Fluorinated Radicals. Rate Constants are for 298 K. 



Radical 


k*i (M 1 s 1 ) 


ftaddJrel) 


/Cm (M- 1 s 1 ) 


h.ti-iM 


RCH2CH2 

(123) a 


1.2 x10 56 


1 


2.4x10 6c 


1 


RCH2CF2 
(124, 77) d 


2.7(±0.5)x10 6e 


22.5 


9.1 (±1.7)x10 6f 


3.8 


ROF2CH2 
(125, 88) 9 


5.2(±1.8)x10 5e 


4.3 


1.4(±0.5)x10 7 ' 


5.8 


RCF 2 CF 2 * 
(99,100)" 


2.0(±0.1)x10 7f 


167 


9.2(±0.8)x10 7f 


38 


RfCH2CH2 
(126,111)' 


1.3(±0.2)x10 5e 


1.1 


2.1 (±0.3)x10 6f 


0.9 


CF 3 CF 2 * 
(118) 


7.9(±1.0)x10 7f 


658 


3.2(±0.3)x10 8f 


133 


C 7 F 15 ' 
(127) 


4.6(±0.6)x10 7y 


383 


2.0(±0.3)x10 8y 


83 



a R = C3H7. * Reference 182, After Modification for Temperature and Other Factors in 
Table III of Reference 183. c Reference 27. d For k aM Experiment, R = C 3 H 7 (124); for 
k H Experiment, R = C 4 H 9 (77). e Reference 88. f Reference 91 and Present Study. 9 For 
k aM Experiment, R = C 3 H 7 (125); for k H Experiment, R = C 4 H 9 (88). " For k aM 
Experiment, R = C 2 H 5 (99); for k H Experiment, R = C 4 H 9 (100). ' For /c add Experiment, 
R f = C 4 F 9 (126); for k H Experiment, R f = C 6 F 13 (111). ' Reference 54. 

systems have been reported. Theoretical studies by Boyd 157 184 at the MP2 level have 
provided reasonably accurate C-H and C-C BDEs for C 2 hydrofluorocarbons. A more 
recent investigation based on isodesmic reactions by Marshall and Schwartz, 158 
published after the completion of the present study, has yielded C-H BDEs of 
appreciably high quality for some linear (C 2 ) and branched (up to C 4 ) polyfluoroalkanes. 
Out of interest in determining additional C-H and C-C BDEs for larger (through C 4 ) 
fluorinated n-alkyls, the geometries of a series of partially fluorinated ethanes, propanes, 
and butanes, along with their respective radicals generated from terminal C-H or C-CH 3 
bond cleavage, were optimized at the hybrid density functional level. This DFT method 
was chosen due to its implicit consideration of electron correlation at only slightly greater 



63 

computational expense than that of Hartree-Fock theory. Utilizing the three-parameter 
exchange functional of Becke 185 and the correlation functional of Lee, Yang, and Parr 186 
with the 6-31 G(d) basis, bond dissociation energies obtained in this "direct" fashion were 
found to be, in the cases where such values are known, within 1-3 kcal mol 1 of those 
determined experimentally and in comparable or better agreement with experiment than 
the MP2/6-311G(d,p) values obtained by Boyd for C 2 systems. Tabulated experimental 
(where known) and calculated C-H and C-C BDE values are provided below in Tables 
3-3 and 3-4, respectively. 

Table 3-3. Theoretical and Experimental C-H Bond Dissociation Energies. 
C-H Bond Calculated BDE, kcal mol' 1 a Experiment 



CH3CH2-H 


100.0" 
97.7 c 
102.0 d 


101.1 ±1 e 


CH3CF2-H 


97.4" 
97.0 c 


99.5 ± 2.5 ' 


CF3CH2-H 


104.3" 
102.0° 
107.1 d 


106.7 ±1 ' 


CF 3 CF 2 -H 


99.5" 
99.7 c 
104.6" 


102.7 ±0.5' 


CH3CH2CH2-H 


100.3" 


100.4 + 0.6 e 


CH3CH2CF2-H 


97.7" 




CH3CF2CH2-H 


103.1 " 




CH3CF2CF2-H 


100.1 " 




CF3CH2CH2-H 


101.4" 




CF3CF2CH2-H 


103.8" 





Reported as D (298.15 K). " B3LYP/6-31G(d); Reference 91 and Present Study. 
c MP2/6-311G(d,p)//MP2/6-31G(d,p); Reference 157. d MP2/6-311+G(3df,2p)//MP2/ 
6-31 G(d); Reference 158. e Reference 187. f Reference 95. 



64 



Table 3-4. Experimental and Theoretical C-C Bond Dissociation Energies. 
C-C Bond Calculated BDE. kcal mol 1 a Experiment 



b nn a j. n n d 



CH3-CH3 89.4 ° 90.4 ± 0.2 

90.6 c 

CF3-CH3 99.6" 101.2 ± 1.1 d 

103.3 c 

CH 3 CH 2 -CH 3 86.3 " 

CH 3 CF 2 -CH 3 91.4" 

CF 3 CH 2 -CH 3 91.4 b 

CF3CF2-CH3 95.5 • 

CH 3 CH 2 CH 2 -CH 3 86.7 ' 

CH 3 CH 2 CF 2 -CH 3 91.6 ft 

CH 3 CF 2 CH 2 -CH 3 89.9 " 

CH 3 CF 2 CF 2 -CH 3 95.4 " 

CF 3 CH 2 CH 2 -CH 3 87.8 * 

3 Reported as D (298.15 K). " B3LYP/6-31G(d); Reference 91 and Present Study. 
MP2/6-311G(d,p)//MP2/6-31G(d,p); Reference 157. d Reference 95. 



c 



Inspection of the data reveals interesting trends within both the C-H and C-C 
BDE series. From Table 3-3, it is observed that a-fluorination results in a weakening of 
C-H bonds, on the order of 1-3 kcal mol" 1 , as predicted by the various levels of theory 
and supported by experiment in the case of ethane versus 1,1-difluoroethane. 
Conversely, p-fluoro substitution results in a 3-5 kcal mol" 1 increase in terminal C-H 
BDEs. Furthermore, strengthening of C-C bonds (Table 3-4) is observed for all systems 
examined, whether substituted at the a, p, or even (albeit diminished) y position. An 
explanation for this behavior, consistent with the given BDE and other thermochemical 
data, is provided later in the chapter. 

With the relative thermodynamics of C-H and C-C bond formation investigated, 
attention is turned to polarity effects. As mentioned previously, rate constants for 



65 

addition of the perfluoroalkyl radicals to subtituted styrenes (Table 2-12) are observed to 
increase with decreasing ionization potential (styrene, IP 8.43 eV; a-methylstyrene, IP 
8.19 eV; pentafluorostyrene, IP 9.20 eV). 188 In contrast, 1,1-difluoropentyl radical (124) 
was found to add with rates equal, within experimental error, for all three olefins. 88 From 
this observation, along with derived absolute electronegativities for the radicals CH 3 * 
(4.96), CH 2 F* (4.73), CHF 2 * (4.91), and CF 3 * (5.74) it was concluded that a-fluoro 
substitution alone does not impart electrophilicity to an alkyl radical, and may instead 
give rise to nucleophilic behavior. 88 

A series of theoretical studies by Wong et a / 4849189190 have assessed the degree 
of charge transfer interaction in the transition states for methyl, hydroxymethyl, 
cyanomethyl and tert-butyl radical additions to a series of monosubstituted olefins. 
Based in part on the computation of partial charges, it was concluded that the addition of 
methyl radical was governed primarily by enthalpic effects, with no evidence for 
nucleophilic character arising from either Mulliken or Bader-based charge-fitting 
schemes. Hydroxymethyl and te/t-butyl were found to be nucleophilic, with cyanomethyl 
exhibiting substantial electrophilicity. 

In order to determine such tendencies for the fluorinated radicals under 
investigation, transition structures for the addition of hydrocarbon and fluoro-substituted 
ethyl radicals to ethylene and d of propene have been located at the UHF/6-31G(d) 
level. Partial atomic charges were then computed, based on fits to the electrostatic 
potential at points selected according the Merz-Kollman-Singh scheme. 191 192 

Previous studies by Houk er a/, utilizing the 3-2 1G basis found that addition of 
ethyl radical to ethylene occurs preferentially via a gauche conformation, with an 
incipient C-C-C-C dihedral angle of ca. 60°. 46 Such gauche and anti structures 
computed at the UHF/6-31(d) level are depicted in Figure 3-18, the former ca. 0.1 
kcal mol" 1 lower in energy than that of (b). Inclusion of electron correlation at the spin- 



66 

projected PMP2/6-311G(d,p)//UHF/6-31G(d) level increases this energy difference to ca. 
0.3kcal mol" 1 . 



-*- 



\ 



103.3° 
103.8° V_\ 

110.0° 2.231 A \ -\ 109.9° 




^ 



2.227 A 




(a) WC # m >~*J (b) 



<^=CS> 



Figure 3-18. UHF/6-31G(d) (a) Gauche and (b) Anti Transition Structures for Addition of 
Ethyl Radical to Ethylene. Relevant Geometrical Parameters are Shown. 



Consistent with UHF/3-21G results, the preferred mode of addition of the ethyl 
and radical to Ct of propene involves a gauche arrangement of the radical with the 
alkene C=C n bond and a transoid orientation of the methyl group of the radical with 
respect to that of the olefinic C 2 carbon (Figure 3-19). This 'gauche-transoid' structure 
lies ca. 0.1 and 0.4 kcal mol" 1 below the 'anti' and 'gauche-cisoid' conformers, 
respectively, at the UHF/6-31(d) level after zero-point energy correction. Inclusion of 
correlation effects (PMP2/6-311G(d,p)//UHF/6-31G(d)) yields energy differences of 0.3 
and 0.4 kcal mol" 1 . These orientation preferences extend to the fluoroethyl series, as 
seen in Tables 3-5 and 3-6. 





\ 2.228 A \ 2.21 9 A 

5^ (c) qfr— ^t 

o 

Figure 3-19. UHF/6-31G(d) (a) 'Gauche-transoid', (b) 'Anti', and (c) 'Gauche-cisoid' 
Transition Structures for Addition of Ethyl Radical to C, of Propene. 




67 



Table 3-5. Geometric Parameters and Total and Relative Energies of Transition 
Structures for Addition of Fluorinated Methyl and Ethyl Radicals to Ethylene. 

Radical r (C-C) r (C=C) z C-C-C ^ &J ZPE E^j; 
(A) (A) (Peg.) (Peg.) (au) (au) 

CH 3 * 2.246 1.382 109.1 101.9 -117.575692 0.089165 

(-118.045817) 

CF 3 * 2.300 1.372 106.6 108.6 -414.156036 0.067955 

(-415.285304) 

CH 3 CH 2 * a 2.227 1.384 110.0 103.8 -156.612999 0.120578 0.0 

(-157.243689) (0.0) 

CH 3 CH 2 * b 2.231 1.383 109.9 103.3 -156.612807 0.120455 0.05 

(-157.243039) (0.34) 

CH 3 CF 2 # a 2.235 1.378 110.0 108.6 -354.333278 0.105597 0.0 

(-355.404529) (0.0) 

CH 3 CF 2 * 6 2.245 1.378 106.3 108.2 -354.332326 0.105470 0.53 

(-355.403222) (0.75) 

CF 3 CH 2 * a 2.258 1.380 109.0 104.1 -453.202939 0.097486 0.0 

(-454.490483) (0.0) 

CFsChV" 2.258 1.380 107.4 103.5 -453.203135 0.097452 -0.14 

(-454.490294) (0.10) 

CF 3 CF 2 * a 2.275 1.375 108.2 108.6 -650.904439 0.081977 0.0 

(-652.632413) (0.0) 

CF 3 CF 2 ,fi 2.280 1.375 104.7 108.1 -650.904343 0.081948 0.04 

(-652.632131) (0.16) 

a Gauche. ' Anti. c Pegree of Radical Pyramidalization. d UHF/6-31G(d) Values; 
PMP2/6-311G(d,p)//UHF/6-31G(d) Values in Parentheses. e Relative Conformer Energy 
Pifferences, kcal mol" 1 . UHF/6-31G(d); PMP2/6-311G(d,p)//UHF/6-31G(d) Values in 
Parentheses. 



Inspection of the data reveals the similarity in both angle of attack and forming 
C-C bond length, regardless of either the nature of the radical or conformation of the 
transition structure. Incipient bond lengths range from ca. 2.22 to 2.30 A, generally 
slightly longer for additions of the fluorinated members of the series. These somewhat 
earlier transition states are also reflected in the shorter olefinic C=C bonds, 1.382 A in 
the case of methyl radical addition to ethylene versus 1 .372 A for trifluoromethyl, in turn 






68 



Table 3-6. Geometric Parameters and Total and Relative Energies of Transition 
Structures for Addition of Fluorinated Methyl and Ethyl Radicals to Ci of Propene. 

Radical r (C-C) r (C=C) Z C-C-C ^ Etotl ZPE E^ 

(A) {A) (Peg.) (Peg.) (au) (au) 

CH 3 * 2.243 1.383 109.3 102.0 -156.614381 0.119634 

(-157.244918) 

CF 3 * 2.297 1.372 106.4 108.9 -453.195586 0.098359 

(-454.485598) 

CH 3 CH 2 * a 2.224 1.385 110.2 103.9 -195.651571 0.150984 0.0 

(-196.442757) (0.0) 

CH 3 CH 2 ,ft 2.219 1.385 111.3 104.2 -195.651075 0.151060 0.35 

(-196.442411) (0.26) 

CH 3 CH 2 * C 2.228 1.384 110.1 103.3 -195.651399 0.150878 0.05 

(-196.442053) (0.38) 

CH 3 CF 2 ,a 2.233 1.379 109.5 108.7 -393.372286 0.135951 0.0 

(-394.604285) (0.0) 

CH 3 CF 2 *" 2.228 1.380 110.7 108.9 -393.371834 0.135994 0.31 

(-394.603661) (0.42) 

CH 3 CF 2 ,C 2.242 1.379 106.0 108.5 -393.371641 0.135892 0.37 

(-394.603448) (0.49) 

CF 3 CH 2 * a 2.254 1.381 109.3 104.2 -492.242374 0.127933 0.0 

(-493.691081) (0.0) 

CF 3 CH 2 * b 2.248 1.381 110.9 104.6 -492.242279 0.128087 0.15 

(-493.691288) (-0.04) 

CF 3 CH 2 ,C 2.257 1.381 107.7 103.6 -492.242269 0.127869 0.03 

(-493.690171) (0.54) 

CF 3 CF 2 * a 2.274 1.375 108.0 108.7 -689.944220 0.112383 0.0 

(-691.833274) (0.0) 

CFaCF/ 6 2.266 1.375 110.1 108.9 -689.944106 0.112473 0.12 

(-691.833236) (0.07) 

CF 3 CF 2 * C 2.279 1.375 104.6 108.3 -689.944068 0.112356 0.08 

(-691.832854) (0.25) 



3 Gauche-transoid. " Gauche-cisoid. c Anti. d Pegree of Radical Pyramidalization. 
e UHF/6-31G(d) Values; PMP2/6-311G(d,p)//UHF/6-31G(d) Values in Parentheses. 
'Relative Conformer Energy Pifferences, kcal mol" 1 . UHF/6-31G(d); PMP2/ 
6-311G(d,p)//UHF/6-31G(d) Values in Parentheses. 






69 



in accord with the greater exothermicity for the latter (AE™ = -22.35 kcal mol" 1 versus 
-34.52 kcal mol" 1 , respectively, at the [QCISD(T)/6-311G(d,p)]7/UHF/6-31(d) level, and 
C-C BDE data in Table 3-4). 

Gauche-transoid addition of ethyl radical to C^ of propene occurs via a transition 
structure with a forming C-C interatomic distance of 2.224 A and and a C=C bond length 
of 1.385 A, in comparison with 2.274 A and 1.375 A for addition of pentafluoroethyl along 
the same trajectory. Attack angles appear slightly smaller for additions of the fluorinated 
radicals, most notably in the case of CH 3 * versus CF 3 * and consistent with a reinforced 
SOMO-HOMO interaction for the latter. However, such differences are barely 
significant, and in the ethyl series appear to be influenced more by the conformation of 
the transition structure than the nature of the attacking radical, in line with the previously 
observed insensitivity of transition state geometry to additions of both nucleophilic and 
electrophilic radicals. 47,48 

Of note is the degree of pyramidalization at the radical site in the addition 
transition structure, ranging from 102-105° for radicals of the RCH 2 * type and 108-109° 
for a-fluorinated species. Considering the pyramidal nature of the ground states of the 
latter as well (Figures 2-9 and 2-10) it follows that a-fluoroalkyl radicals enjoy a kinetic 
advantage over their hydrocarbon analogues in that little or no additional bending is 
necessary to accommodate the addition transition structure. The energetic cost of 
pyramidalization of the methyl and fert-butyl radicals to the same extent as required for 
their addition to ethylene has been computed at 1.5 and 1.6 kcal mol" 1 , respectively, at 
the RMP2/6-31G(d)//UHF/6-31G(d) level. 49 

Calculated degrees of charge transfer (CT) between radical and olefin moieties 
of the addition transition structures are provided in Table 3-7. Where applicable, lowest 
energy conformations (gauche in the case of ethyl radical additions to ethylene, gauche- 



70 

transoid for additions to Ct of propene) were used in the determination of the 
electrostatic potential-derived charges. 



Table 3-7. Calculated Charge Transfer Data (Electrons) for Transition Structures of 
Hydrocarbon and Fluorinated Methyl and Ethyl Radical Addition to Ethylene and Ci of 
Propene. 



Radical 


Ethylene a 


Propene b 


CH 3 * 


-0.019 


-0.004 


CF 3 * 


-0.013 


-0.006 


CH3CH2 


+0.036 


+0.052 


CH3CF2 


+0.030 


+0.052 


CF3CH2 


-0.047 


-0.037 


CF 3 CF 2 * 


-0.045 


-0.034 



Note: Derived from UHF/6-31G(d) Electrostatic Potentials. Negative Values Denote 
Electron Transfer from Alkene to Radical. a Gauche Transition Structure. " Gauche- 
transoid Transition Structure. 

Consistent with previous investigation, addition of CH 3 * to ethylene involves only 
a slight degree of charge transfer (-0.019 e) from olefin to radical (Mulliken and Bader 
analyses yield values of -0.017 and -0.011, respectively) 48 and even less so for addition 
to propene. Somewhat surprisingly, CF 3 * addition is also predicted to occur without 
appreciable polarization. 

Along the ethyl series, such interactions appear more clearly defined. Ethyl 
radical addition to both ethylene and propene involves a shift of electron density from the 
radical to the alkene (0.036 and 0.052 e, respectively) well in accord with the expected 
nucleophilicity of the alkyl radicals. CH 3 CF 2 ' is predicted to be nucleophilic as well, 
exhibiting transition state polar characteristics very similar to those of its hydrocarbon 
counterpart and consistent with the experimentally deduced non-electrophilicity of 
the a-fluoro radicals. 



71 

A striking reversal in these trends occurs upon fluorination at the p carbon atom, 
regardless of the nature of the radical site itself. Here it is seen that both CF 3 CH 2 * and 
CF3CF2* exhibit substantial electrophilicty, with ca. 0.04 - 0.05 units of electron density 
transferred from the alkene to the radical center. To place such values into some 
degree of perspective, addition of the strongly electrophilic cyanomethyl radical to C 2 of 
electron-rich vinylamine is predicted to occur with a transfer of ca. 0.11 electrons from 
CH 2 =CHNH 2 to *CH 2 CN. 48 

It is especially noteworthy that the degree of CT in the case of CF 3 CH 2 * and 
CF 3 CF 2 * addition is practically unaffected by fluorination at the a carbon (-0.047 versus 
-0.045 and -0.037 versus -0.034, respectively). This, along with the demonstrated lack 
of kinetic impact of fluorine substitution at the y position (Table 3-2) leads to the 
conclusion that the electrophilic character of the perfluoroalkyl radicals derives 
exclusively from substitution at the 2-position. 

With geometric, enthalpic, and polar considerations for hydrofluorocarbon 
radicals having been addressed, the influences of each of these effects on determined 
/(add and k H values are now discussed. 

a,g-Difluoroalkvl Radicals (77. 124) 

The 1,1-difluoroalkyl radicals, as mentioned previously, benefit from 
pyramidalization at their radical site, leading to a more facile adoption of the transition 
structure for addition or hydrogen atom abstraction. However, such an advantage is 
counteracted by the experimentally and theoretically demonstrated lack of electrophilicity 
for such species. Moreover, the terminal C-H bond weakening effect of gem-difluoro 
substitution (2 - 3 kcal mol 1 , Table 3-3) leads to a slight thermodynamic disadvantage 
for hydrogen abstraction by the corresponding radical relative to the hydrocarbon. Thus, 
the 3.8-fold rate enhancement enjoyed by 77 may be completely ascribed to the a-type, 



72 

pyramidal nature of its ground state, attenuated by enthalpic and polarity factors working 
in opposition. In contrast, the strengthening effect of gem-difluorination on C-C bonds 
compliments that of pyramidal geometry, leading to a more substantial (22.5-fold) rate 
enhancement for the addition of 124 to styrene versus hydrocarbon 123. 

B,B-Difluoroalkvl Radicals (88. 125) 

The rate enhancements for addition (4.3) and hydrogen abstraction (5.8) 
observed for 2,2-difluoroalk-1-yl radicals are due to a complimentary combination of 
polar and enthalpic effects. The terminal C-H bond in 2,2-difluoropropane is predicted to 
be 2.8 kcal mol" 1 stronger than that of propane itself; similarly, gem-difluorination at C 2 
of butane leads to a 3.2 kcal mol" 1 strengthening of its C 3 -C 4 bond. In spite of these 
favorable considerations, the near-planar ground state geometry of 88 results in a 
modest net rate acceleration for hydrogen abstraction from nBu 3 SnH. This also 
functions to oppose the CT and enthalpic advantages present in the addition reaction of 
125, giving rise to only a slight rate increase relative to n-alkyls and certainly diminished 
in comparison with that enjoyed by 124. 

Y-Fluorinated Radicals (111. 126) 

Due to the near-planar geometric character of 3-fluoroalk-1-yls and the lack of 
effect (ca. 1 kcal mol" 1 ) on terminal C-H and C-C BDE values, fluorination beyond two 
carbon atoms removed from the radical site exhibits a negligible effect on the rates of 
both addition and hydrogen transfer. The reactivities of 111 and 126 are found to be, 
within experimental error, identical to those of the corresponding hydrocarbon. 

g.g.B.B-Tetrafluoroalkvl Radicals (99,100) and Pentafluoroethyl Radical (118) 

Radicals substituted at the a and p positions benefit from both pyramidal 
geometries and electrophilic character. Since the degree of pyramidalization of 1,1- 



73 

difluoroalkyl radicals remains constant regardless of substitution at the p- and further 
positions, geometrically induced influences on the reactivities of such polyfluorinated 
radicals are expected to be uniform. Consequently, rate enhancements for radicals of 
the type RCH 2 CF 2 CF 2 *, R,CF 2 CF 2 CF 2 *, and CF 3 CF 2 ' versus RCH 2 CH 2 CF 2 * (for k H : 38, 
83, and 133 versus 3.8; for k aM : 167, 383 and 658 versus 22.5; all relative to n-alkyls) 
derive from either an increasing degree of transition state charge transfer stabilization, 
increasingly greater exothermicity of reaction, or a combination of both. The relevant 
C-H BDE data in Table 3-3 yields no direct correlation between reaction rate and 
enthalpy for the polyfluorinated radical series, with values of 97.7, 100.1, and 99.5 
kcal mol" 1 corresponding to hydrogen abstraction by radicals 77, 100, and 118. 
Similarly, terminal C-C BDEs of 91.6, 95.4, and 95.5 kcal mol" 1 , equated with the 
additions of 124, 99, and 118, illustrate that although p-fluoro substitution should lead to 
rate enhancement on thermochemical grounds, such an effect does not account for the 
incremental acceleration across the series. 

With the degree of radical electrophilicity related to its substitution at the 2- 
position and the potential for additional p C-F a* derealization made possible by the 
"extra" fluorine substituent in CF 3 CF 2 * relative to RCH 2 CF 2 CF 2 * and R,CF 2 CF 2 CF 2 *, it 
follows that the increasing resonance and inductive withdrawal ability of these groups 
relative to RCH 2 CH 2 CF 2 ' sufficiently explain both the enhanced reactivity of these 
radicals as a whole, as well as the observed trend. 

Fluorine Substituent Effects on Bond Dissociation Energies; Coulombic Interactions 

As previously discussed, substitution by fluorine in hydrocarbons gives rise to 
nearly additive and sometimes opposite effects on C-H and C-C homolytic bond 
dissociation energies. For example, the aforementioned 1-3 kcal mol" 1 weakening 
effect of a,a-difluoro substitution and the 3-5 kcal mol" 1 strengthening of terminal C-H 



74 

bonds by p-fluorination lead to near cancellation in the case of pentafluoroethane and 
1,1,2,2-tetrafluoropropane, yielding BDE values very near those of the parent 
hydrocarbon (Table 3-3.) Furthermore, the 4.9 kcal mol' 1 strengthening brought about 
by substitution at the breaking C-C bond is reinforced by an additional 3-5 kcal mol" 1 
upon further fluorine incorporation at the p position, leading to net increases of nearly 10 
kcal mol" 1 over the parent in the C 2 -C 3 homolysis of 1,1,1,2,2-pentafluoropropane and 
cleavage of the terminal C-C bond of 2,2,3,3-tetrafluorobutane. 

The opposite effects of a,a-difluoro substitution on C-H and C-C bond 
dissociation energies bear special mention. Experimental BDE values in the 
fluoromethanes are in accord with the general RSE expectations of Pasto (Table 2-7) in 
that although substitution at a radical center by a single fluorine is stabilizing, its further 
incorporation leads to a successive decrease in RSE, resulting in net destabilization for 
CF 3 *. This is consistent with the incremental strengthening of C-F bonds along the 
fluoromethane series, leading to a C-H BDE in CF 3 H which is 1.9 kcal mol" 1 stronger 
than that of methane itself (methane BDE, 104.8 kcal mol" 1 ; see discussion below Table 
2-2) and the comparatively weaker C-H bonds in CH 3 F and CH 2 F 2 . 

The considerable stability of the 2,2-difluoroalkanes relative to their 1,1-difluoro 
isomers, demonstrated by the isodesmic reaction in Equation 3-2 (calculated from 
B3LYP/6-31G(d) total energies and zero-point corrections) provides the underlying 
reason for why the stability trends observed above do not extend to C-C bonds. 

CH 3 CF 2 CH 3 + CH3CH3 ■ CH 3 CF 2 H + CH3CH2CH3 (3-2) 

AErxn = + 7.8 kcal mol" 1 

In Chapter 2, the Wiberg rationale of electrostatic attraction for the incremental 
strengthening and shortening of C-F bonds in the fluoromethanes was introduced. 
Similarly, it is found that such a Coulombic-based argument sufficiently explains the 
observed effects of fluorine substitution on C-H and C-C bond dissociation energies. 



75 

Atomic charges for select hydrofluorocarbons based on the B3LYP/6-31G(d) 
electrostatic potential (Merz-Kollman radii) are provided in Table 3-8. 

The dipolar nature of the C-C bond in 1,1,1-trifluoroethane and its resultant 
increase in BDE relative to ethane and hexafluoroethane (Table 2-3) was first postulated 
by Rodgers. 156 Such stabilization due to increased C-C bond ionicity is seen in 
Equations 3-3 (derived from experimental heats of formation 188 ) and 3-4 (from B3LYP/6- 
31G(d) total and zero-point energies). 

Table 3-8. Atomic Charges in Hydrofluorocarbons, Based on B3LYP/6-31G(d) Density. 

Hydrofluorocarbon xE x£ a xQe xH a xhb 

C a H 3 C a H 3 -0.055 +0.018 

Ethane 

C a H 2 F 2 -0.200 +0.320 +0.041 

Difluoromethane 

CpH 3 C a F 2 H -0.228 +0.467 -0.386 +0.020 +0.110 (2H) 

1 , 1 -Difluoroethane +0. 135 (1 H) 

C„H 3 C a F 2 C p H 3 -0.245 +0.631 -0.477 +0.133 (4H) 

2,2-Difluoropropane +0.140 (2H) 

CF3CF3 + CH3CH3 - 2 CH3CF3 (3-3) 

AE rxn = -16.9kcal mol" 1 

CH 3 CH 2 CH 2 CH3 + CH 3 CF 2 CF 2 CH3 - - 2 CH 3 CF 2 CH 2 CH 3 (3-4) 

AErxp = -5.0 kcal mol" 1 

The significant electrostatic attraction between adjacent carbon atoms in both 
CH 3 CF 2 H and CH 3 CF 2 CH 3 is readily apparent from the data in Table 3-8, providing an 
explanation for the strengthening of these bonds relative to their hydrocarbon or 
perfluorocarbon analogues. In addition, C-H repulsion in difluoromethane and 1,1- 
difluoroethane is predicted, in accord with the observed weakening of these bonds 
compared to those of methane and ethane. Conversely, the strong attraction between 



76 

the p carbon and hydrogen atoms of CH 3 CF 2 H faC, -0.386; xH av g, +0.118) and 
CH 3 CF 2 CH 3 (xC, -0.477; x H avg, +0.135) is consistent with their greater theoretical and 
experimental BDEs. 

Conclusion 

Based on time-resolved /c a <jd measurements, absolute rate constants for 
hydrogen abstraction from tri-n-butyltin hydride have been determined for a series 
partially fluorinated radicals. The reactivities of such radicals towards nBu 3 SnH follow 
those of addition to alkenes. The enhanced reactivity of oc.cc-difluoroalkyl radicals in 
hydrogen abstraction reactions derives exclusively from their pyramidal geometry. 
P-Fluorination leads to a favorable combination of polar and thermodynamic factors in 
both addition and hydrogen transfer reactions, giving rise to the exceptional reactivity of 
CF3CF2* and the perfluoroalkyl radicals as a whole. In Chapter 4, the k H values so 
obtained are utilized in the determination of absolute rates of cyclization for partially 
fluorinated 5-hexenyl radicals. 

A self-consistent rationale for the impact of fluorine substitution on C-H and C-C 
bond dissociation energies based on electrostatic considerations was offered, providing 
new understanding of the thermochemistry of bonding and radical stabilization in 
hydrofluorocarbons. 






CHAPTER 4 

THE REACTIVITY OF PARTIALLY FLUORINATED RADICALS 
IN INTRAMOLECULAR CYCLIZATION REACTIONS 



Introduction 

The intramolecular addition reactions of 5-hexen-1-yl radicals continue to attract 
the attention of synthetic and physical organic chemists alike. Such cyclizations to 
(predominantly) 5-exo products have been utilized as probes for the detection of radical 
intermediates and as basis reactions for the competitive determination of absolute 
kinetic data for a number of free radical transformations. 16 Rationalization of the rates, 
and especially the regio- and stereochemistry, of intramolecular radical additions on the 
basis of force field 63646667 and molecular orbital 62 ' 65,66 ' 68 techniques has proven to be one 
of the greatest successes of theory in the prediction of organic reactivity. Due in no 
small part to such structure-reactivity studies, application of free radical methodology to 
the singular and tandem construction of 5-membered rings has been equally exploited, 
providing for the assembly of functionalized organic systems under mild conditions, often 
accomplished with a high degree of stereocontrol. 11 " 15 

Determination of absolute rates of cyclization of per- and other highly fluorinated 
5-hexenyl systems 8689,170 have aided in solidifying the understanding of the effect of 
fluorine substitution on the reactivity of organic radicals, though at the same time 
generating a number of new questions, particularly with regard to cyclization 
regiochemistry. 

In order to examine the potentially more subtle influences of partial fluorination 
on 5-hexenyl radical reactivity, and to obtain a set of data through which the effect of 



77 



78 

incremental gem-difluoro substitution along the aliphatic portion of the 5-hexenyl chain 
may be assessed, absolute rates of 5-exo and 6-endo cyclization for some partially 
fluorinated 5-hexenyl radicals have been determined based on competitive kinetic 
technique and the absolute rates of hydrogen abstraction obtained in Chapter 3. 

Precursor Syntheses and Competitive Kinetic Studies 

As in the bimolecular addition versus hydrogen abstraction competition studies, 
bromide precursors were utilized in the generation of partially fluorinated 5-hexenyl 
radicals. Photolysis by UV irradiation (Rayonet photoreactor) in the presence of known, 
varying concentrations of hydrogen atom donor, carefully adjusted to ensure pseudo-first 
order kinetic behavior and to allow for accurately measurable amounts of cyclization and 
hydrogen abstraction products, provided the kinetic ratio k H I k Cn - Absolute rate 
constants for 5-exo and (where applicable) 6-endo cyclization were then determined 
from the known value of hydrogen abstraction rate constant k H , illustrated in Figure 1-18 
and in greater detail below. 

1.1-Difluorohex-5-en-1-yl Radical (128) 

Synthesis of bromide 135 was achieved in six steps in ca. 14.5% overall yield, 
starting from commercially available 3-buten-1-ol (129, Figure 4-1). Curiously, direct 
addition of dibromodifluoromethane to 129 could not be induced, even through extended 
reaction time at elevated temperatures. Although the presence of the alcohol 
functionality in the alkene starting material would not have been expected to exhibit a 
detrimental effect (in light of the hydroxylic nature of the ethanolamine / te/t-butanol 
cosolvent medium) protection of the hydroxyl moiety as its tert-butyldimethylsilyl ether 
130 (TBDMSCI, imidazole in dimethylformamide) followed by dibromodifluoromethane 
addition indeed afforded 1,3-dibromo-1,1-difluoro adduct 131 in good yield. Highly 
selective displacement of the internal bromine yielded bromodifluoromethyl derivative 






79 



/V>! 



TBMSCI 



OH 



ImH, DMF 



129 



OTBDMS 
(89.3%) 
130 



CF 2 Br 2 



tBuOH, H 2 NCH 2 CH 2 OH 
CuCI (cat.) 



Br 



BrF 2 C 



NaBH. 



OTBDMS 



DMSO 



(72.4%) 
131 



BrF 2 C 



OTBDMS 



(90.0%) 
132 



FeCI, 



CH 3 CN 



BrF 2 C 



(97.7%) 
133 



PCC 



OH CH 2 CI 2 



BrF 2 C^^A H 



(53.1%) 
134 



Ph 3 P=CH 2 
THF 



CF 2 Br 



(48.1%) 
135 

Figure 4-1. Preparation of 6-Bromo-6,6-difluorohex-1-ene, Precursor to 1,1-Difluorohex- 
5-en-1-yl Radical 128. 

132 with virtually no overreduction product, as monitored via 19 F NMR through high 
conversion of starting material. Lewis acid deprotection via the method of Cort 193 and 
subsequent pyridinium chlorochromate oxidation provided aldehyde 134, further 
subjected to Wittig olefination to yield precursor 135. 

Generation of 128 was achieved via irradiation of a solution of 135 in C 6 D 6 in the 
presence of excess nBu 3 SnH (Figure 4-2) and an internal standard of a,a,ct- 
trifluorotoluene. Direct capture of 128 by hydrogen atom donor afforded reduction 
product 6,6-difluorohex-1-ene 136, whereas intermediate 137, subsequently trapped by 
nBu 3 SnH to yield spectroscopically observable cyclization product 138, was generated 
via irreversible, unimolecular rearrangement with rate constant k C 5 (no 6-endo cyclization 
was observed, within NMR detection limits, (ca. 4%) for 128). During the course of the 









80 





















/\ /v. CFoBr 


hv 




/\/V CF2 ' 








#\/\/ 




nBu 3 SnH 






135 




C 6 D 6 




128 










nBu 3 SnH 




136 


nBu 3 Sn' 

u 3 SnH 




/ 


*H 


*" < 




\ 


f 6<: ] 


nB 


M 




^C5 






► 


WV 








137 






138 
















+ 


nBu3Sn* 



Figure 4-2. /( H / /(c Competitive Kinetic Scheme for 1,1-Difluorohex-5-en-1-yl Radical 
128 



reaction, tributylstannyl radicals generated by transfer of hydrogen atom from nBu 3 SnH 
to 128 and 137 served to propagate the chain process via bromine abstraction from 135. 
Product ratios for varied concentrations of nBu 3 SnH were determined by 19 F 
NMR analysis according to the pseudo-first-order relation in Equation 4-1, 



[136] 
[138] 



[^h] [128] [nBu 3 SnH] 
[kcs] [128] 



(4-1) 



a plot of which obtained for each data point versus nBu 3 SnH concentration providing the 
ratio k H I k C5 - Exceptionally clean spectra and high mass balances were obtained for 
each kinetic point, indicating the efficiency of the radical chain process and reliability of 
the obtained rate constant ratios. A partial 19 F spectrum of the first of six data points is 
provided in Figure 4-3, a doublet of triplets (-CF 2 H, <|> -116.2) observed for 136 versus 
overlapping doublets of doublets of triplets at f -100.3 and -107.8 for the diastereotopic 
-CF 2 - resonances of 138. Kinetic data and product yields are given in Table 4-1, a plot 
of which found in Figure 4-4. The slope of the line (2.57 ± 0.05) in conjunction with the 



if 



81 







-96 



-98 



-100 -102 -104 -106 



•108 



■110 



-112 



■114 



-116 



-118 



Figure 4-3. Partial 19 F NMR Spectrum of Data Point 1 for k H I k c Competition of 1,1- 
Difluorohex-5-en-1-yl Radical 128. 



Table 4-1. Competitive Kinetic Data for k H I k c Competition of 1,1-Difluorohex-5-en-1-yl 
Radical 77. 

r 135 1 LnBu-jSnH] \ 136 l/f 1381 % Yield 



0.054 


0.673 


1.53 


88 


0.054 


0.807 


1.91 


100 


0.054 


0.942 


2.28 


89 


0.054 


1.08 


2.57 


94 


0.054 


1.21 


2.93 


95 


0.054 


1.35 


3.29 


92 



00 
CO 



CO 




0.6 



0.7 



0.8 



0.9 1.0 1.1 

[ nBu 3 SnH ] 

Figure 4-4. Plot of the Data in Columns 2 and 3 of Table 4-1 . 



1.2 



1.3 



1.4 



82 

known absolute rate constant for hydrogen atom abstraction from nBu 3 SnH by 1,1- 
difluorohex-1-yl radical 77, 9.1 (± 1.7) x 10 6 M" 1 s" 1 , resulted in a k C5 value of 
3.5 (± 0.59) x 10 6 s" 1 for 5-exo closure of 128, with errors in k c reflecting both the least- 
squares fit of the line and propagated error in k H . Syntheses of hydrogen atom transfer 
and cyclization products 136 and 138 were performed as shown in Figure 4-5. 



nBu 3 SnH 
CF 2 Br ^ . . CF 2 H 



AIBN 
135 Mesitylene 136 



DAST A F 





C/H2CI2 

139 138 

Figure 4-5. Preparation of Hydrogen Abstraction and 5-Exo Cyclization Products 136 
and 138. 



2,2-Difluorohex-5-en-1-vl Radical (140) 

Bromide 144 was obtained in a three-step synthesis starting from 1,2-epoxy-5- 
hexene (Figure 4-6). Regiospecific ring opening by a Corey 194 procedure afforded 
bromohydrin 142, converted to the corresponding oc-haloketone via Jones oxidation. 
Treatment of 143 with DAST in dichloromethane afforded precursor 144 in 40.9% overall 
yield, purified by preparative GC for competitive kinetic study. 

In contrast to the virtually regiospecific 5-exo closure of 128, a broad singlet 
resonance at (j> -95.8, comprising approximately 9% of cyclized products, was observed 
in the 19 F NMR spectra for the k H I k c competition of 140. This is attributed to competing 
6-endo cyclization to 148 (Figure 4-7), the presence of which was confirmed by spectral 
comparison with that of an authentic sample of 149. 






83 






^\/\A 



141 



KBr, CH 3 C0 2 H 


^-^Br 
OH 


Na 2 Cr 2 7 / H 2 S0 4 


W 


THF/H 2 




Et 2 






(88.1%) 










142 









DAST 



^ 



^~ 



CHzBr CH 2 CI 2 



CF 2 CH 2 Br 



(82.6%) 
143 



(56.2%) 
144 



Figure 4-6. Preparation of 6-Bromo-5,5-difluorohex-1-ene, Precursor to 2,2-Difluorohex- 
5-en-1-yl Radical 140. 



^/v^CFzCHaBr 



hv 



144 



nBu 3 SnH 
C 6 D 6 



-^^ N ^CF 2 CH 2 
140 



nBu 3 SnH 



\X^CF 2 CH 3 
145 



<C5 




F 
146 



nBu 3 SnH 




F 
147 



<C6 



F. F 




F x F 



nBu 3 SnH 




148 



149 



Figure 4-7. k H I k c Competitive Kinetic Scheme for 2,2-Difluorohex-5-en-1-yl Radical 
140 



84 

Cyclizations of p,p-difluoroalkyl radicals have appeared in the synthetic 
literature, 195 utilized in the generation of alkoxy-substituted gem-difluorocyclopentane, 
cyclohexane, and tetrahydropyran derivatives, though reported to undergo addition in an 
exo-specific manner. However, the regiochemical behavior exhibited in the cyclization of 
140 provided experimental verification of that previously predicted on the basis of ab 
initio calculations, performed as part of the present study and elaborated upon in the 
Discussion section of the chapter. Competition plots for k H I k C 5 and k H I k C e are found in 
Figures 4-8 and 4-9, respectively. Preparation of hydrogen abstraction and 5-exo and 6- 
endo cyclization products was performed as shown in Figure 4-10. 




0.25 



0.30 



0.35 



0.50 



0.55 



0.40 0.45 

[ nBu 3 SnH ] 
Figure 4-8. Plot of k H I k C5 Competition of 2,2-Difluorohex-5-en-1-yl Radical 140. 



0.60 





60 - 


Coefficients: 








50 - 


m = 132 
b =-14.0 




m^-^"^ 




40 - 
30 - 


r 2 = 0.995 










I 


I I 


I I I 



0.25 



0.30 



0.35 



0.50 



0.55 



0.40 0.45 

[ nBu 3 SnH ] 
Figure 4-9. Plot of k H I k C6 Competition of 2,2-Difluorohex-5-en-1-yl Radical 140. 



0.60 



85 



„«... nBu 3 SnH 
^^ X/ CF 2 CH 2 Br ■ ^ ^.CFjCHa 

AIBN 
144 Mesitylene 145 



DAST 




OH2WI2 




150 147 




R F 




DAST ^ 

CH2CI2 



151 149 

Figure 4-10. Preparation of Hydrogen Abstraction and 5-Exo and 6-Endo Cyclization 
Products 145, 147, and 149. 



1,1,2,2-Tetrafluorohex-5-en-1-vl Radical (152) 

Bromide 101 (Halocarbons, Inc.) served as the precursor to a,a,p,p- 
tetrafluorinated radical 152. However, attempts at determination of accurate k H I k c 
ratios using nBu 3 SnH as a trapping agent met with failure, leading primarily to reduction 
product 153, with only minor amounts of 155 and 177 evident in the 19 F NMR baseline 
which could not be integrated accurately over a span of hydrogen atom donor 
concentrations (Figure 4-11). In principle, lowering the concentration of both radical 
precursor (typically in the 0.05 - 0.1 M range) and trapping agent (while still maintaining 
pseudo-first order conditions) should effectively decrease the amount of reduction 
product and allow for a greater degree of cyclization to be observed. However, too great 
of a decrease in precursor concentration leads to decreased NMR signal to noise ratios, 
the necessity of longer acquisition times per sample, and increased potential for the 
introduction of systematic error. 






86 



XF 2 Br 



101 



nBu 3 SnH 



hv 

nBu 3 SnH 
C 6 D 6 





CF 2 H 



<C5 



virtually no cyclization products 
observed 



k C& 

Figure 4-11. Attempted k H I k c Competition of 1,1,2,2-Tetrafluorohex-5-en-1-yl Radical 
152 with nBu 3 SnH as Trapping Agent. 



As it was evident that any cyclization reaction of 152 occurred with a rate 
constant too low to be competitive with transfer of hydrogen from nBu 3 SnH, attention 
was turned to alternative trapping agents. With the rate of hydrogen atom transfer to 
perfluoroalkyl radicals by a number of reducing agents having been accurately 
determined (Table 2-10), it was decided to investigate the suitability of 
tris(trimethylsilyl)silane ((TMS) 3 SiH) as a competitive trapping agent for the calibration of 
cyclization rate constants for 152, due to its approximately four-fold decrease in 
hydrogen transfer rate to perfluoroalkyls relative to nBu 3 SnH. For such a competition to 
be of kinetic value, however, it was necessary to determine rate constant k H for 
tetrafluoroalkyl radical 100 with (TMS) 3 SiH, using its known rate of addition to styrene as 
a competing basis reaction. The plot for the k H ((TMS) 3 SiH) / /c add (styrene) competition 
of 100 is provided in Figure 4-12. 









87 



With the k H ((TMS) 3 SiH) value of 1.8 (+ 0.1) x 10 7 M' 1 s" 1 for 100 in hand (which, 
along with its k H (nBu 3 SnH) of 9.2 (± 0.8) x 10 7 M" 1 s" 1 , (Table 3-2) may be compared 
with 5.1 x 10 7 M" 1 s" 1 and 2.0 x 10 8 M" 1 s" 1 , respectively, for perfluoro-n-alkyl radicals; 
Table 2-10) rate constants k Cb and k C6 for 152 were then determined (Figures 4-13 and 
4-14.) Use of this slower hydrogen transfer agent allowed for sufficient competitive 
(including significant 6-endo) cyclization such that accurate k H I k Cn ratios could be 
obtained. Isolation of products 153, 155, and 157 was achieved by slow syringe pump 
addition of nBu 3 SnH to a heated, irradiated solution of 101 in mesitylene (Figure 4-15). 



00 

o 




0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 
[ (TMS) 3 SiH ] / [ C 6 H 5 CH=CH 2 ] 

Figure 4-12. Plot for k H ((TMS) 3 SiH) / k aM (Styrene) Competition of 1,1,2,2- 
Tetrafluorohex-1-yl Radical 100. 



in 


£..0 

2.0 - 


Coefficients: ^— —-**" 
m = 2.11 ^ — 




b =-0.242 ^~^^^*^ 


2 


1.5 - 


r' = 0.998 ^^^-^^^ 


CO 

to 


1.0 - 
n ^ 


^^^-*^^ 






I I I I I I 



0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 

[ (TMS) 3 SiH ] 

Figure 4-13. Plot of k H I k C5 Competition of 1,1,2,2-Tetrafluorohex-5-en-1-yl Radical 152. 



88 



10 



CO 

m 



10 
9 

8 - 
7 

6 - 
5 - 



Coefficients 
m = 9.44 
b =-1.15 
' 0.999 




[ (TMS) 3 SiH ] 
Figure 4-14. Plot of k H I k C6 Competition of 1 ,1 ,2,2-Tetrafluorohex-5-en-1-yl Radical 152. 



,CF 2 Br 



101 



nBu 3 SnH 



AIBN 
Mesitylene 



XF, 



F 2 
152 



nBu 3 SnH 



,CF 2 H 



153 




F 
154 

R F 




156 



nBu 3 SnH 



nBu 3 SnH 




F 
155 

K F 




157 



Figure 4-15. Preparation of Hydrogen Abstraction and 5-Exo and 6-Endo Cyclization 
Products 153, 155, and 157. 



89 
Discussion 

Absolute rate constants of cyclization for radicals 128, 140, and 152 are given in 
Table 4-2. For comparison, such k C5 and (where applicable) k C6 values for parent 
hydrocarbon 1 and fluorinated radicals 65 and 68 are also provided, the latter two 
systems along with those of the current study found to give rise to the greatest impact on 
cyclization kinetics and regiochemistry. Recent studies of 5-hexenyl systems bearing 
vinylic fluorine substituents have demonstrated that the effect of such substitution is 
relatively minor, with no 6-endo products observed within the detection limits imposed by 
NMR analysis and kc*w values with respect to 1 ranging from ca. 0.09 to 2.3. 170 

Cyclization Kinetics 

Rates of intramolecular addition of partially-fluorinated radicals to alkenes should 
be governed by the same combination of steric, polar, and thermodynamic factors which 
influence the reactivity of their intermolecular counterparts. As seen by comparison of 
the data in Tables 3-2 and 4-2, the reactivity characteristics of the above radicals in 
unimolecular cyclization reactions, particularly 5-exo closure, generally reflect those 
observed in bimolecular additions. This is logical in light of the similarity of their 
transition structures, elaborated upon in Chapters 1 and 3. 66 

The pyramidal nature of a,a-difluoroalkyl radicals, combined with the more 
favorable thermodynamics of C-C bond formation involving fluorinated carbon (see 
related discussions in Chapters 2 and 3, along with cyclization transition structures and 
energies of reaction below) provide sufficient explanation for the 13-fold increase in rate 
of 5-exo ring closure of 128 relative to 1. The factor of 22.5 observed for addition of 77 
versus 124 to styrene is consistent with the observed cyclization rate ratios. 

The increase in k C s of 4.1 enjoyed by 140 parallels that of bimolecular addition of 
125 to styrene, (4.3) due to its increased electrophilicty over both hydrocarbon 1 and 



90 



Table 4-2. Absolute Rate Constants for 5-Exo and 6-Endo Cyclization of Partially 
Fluorinated 5-Hexenyl Radicals. Rate Constants are for 303 K; Relative k Cn Values in 
Parentheses. 



Cyclization Reaction 





^C5. 


10 5 s" 1 


*C6 


10 5 s" 1 




2.7 a 




0.05 a 




(1) 




(1) 




CF 2 



128 



CF, + 



137 



^CF 2 



158 



35 (± 5.9) b N/A b ' c 

(13.0) 



XF, 



140 



CF 2 
146 




C 
F 2 

148 



1 1 (± 3.8) 
(4.1) 



1.1 (±0.34) 
(22) 



CF, 

i 

XF 2 



152 



CF 2 + 
CF 2 



154 



C C .CF 2 
F 2 
156 



87 (±4.1) 
(32.2) 



19 (±1.1) 
(380) 



C 
F 2 



65 



CF 2 
.CF 2 



CF, + 



F 2 C-CF 2 
66 



F 2 C. C .CF 2 
F 2 
67 



440 (i 46) d 52 (±6.4) d 



(163) 



(1040) 



F 2 C. C .CF 2 
F 2 

68 



F 2 C CF 2 + 
F 2 C-CF 2 

69 



F 2 C C .CF 2 
F 2 



110 (±1.7) d 35(±4.4) d 
(40.7) (700) 



70 



3 Reference 16. b Current Study. c 6-Endo Cyclization Not Observed Within 19 F NMR 
Detection Limits (Approximately 4%). d Reference 170. 






91 

a.a-difluorocarbon 128 and greater exothermicity of addition relative to n-alkyls, though 
tempered by the effectively planar, rc-nature of its radical center. 

a,a,p,p-Tetrafluorinated radical 152, as in the case of bimolecular additions, 
benefits from favorable thermodynamics of addition as well as its electrophilicty and 
a-character, leading to the 32-fold increase in /c C s compared to parent 1. It should be 
noted that such unimolecular cyclizations possess an inherent entropic advantage over 
their bimolecular analogues, generally proceeding with log A values ca. 2 units larger 
than those for the latter 16 and resulting in a leveling of rate ratios relative to 
intermolecular additions. 

Upon additional fluorination of the aliphatic moiety of the 5-hexenyl chain (65, 68) 
such radicals undertake perfluoroalkyl character, leading to further increase in reactivity 
akin to that observed for C 7 F 15 * (127, k aM m = 383) versus CH 3 CH 2 CF 2 CF2* (99, 
kaM <rei) = 167, relative to n-alkyl) in bimolecular additions to styrene. Geminal 
difluorination at the allylic position (68) serves to diminish the transition state SOMO- 
HOMO interaction, and hence k C 5 and /c C6 , relative to 65. 



Cvclization Reqiochemistry 






The significant degree to which 152, 65, 68, and even 140 undergo 6-endo 
cyclization is particularly striking, with six-membered ring formation in 140 occuring with 
a rate nearly half, and 65 and 68 more than an order of magnitude greater than, that of 
5-exo closure for hydrocarbon 1. In comparison, 5-hexenyl systems bearing alkyl 
substituents along the aliphatic fragment exhibit regiochemical profiles similar to that of 
the unsubstituted parent. 7071 

The question of potential reversibility in the above cyclizations has been 
addressed, in light of the greater relative thermodynamic stability of secondary 
cyclohexyl radicals. Upon independent generation of 5-exo adduct radical 69 from 
precursor 1-(iodomethyl)-2,2,3,3,4,4,5,5-octafluorocyclopentane in the presence of 



92 

hydrogen atom donor triethylsilane in C 6 D 6 , the only product observed after complete 
consumption of starting material was that resulting from direct capture of 69 by Et 3 SiH. 170 
The lack of 6-endo or ring-opened products originating from 69, coupled with the ab initio 
predictions based on relative energies of cyclization transition structures described 
below, demonstrates that the regiochemical characteristics of fluorinated 5-hexenyl 
radical cyclizations are indeed kinetic in nature. 

Of further note is that system 68, which undergoes the greatest percentage 
(24.1%) of 6-endo closure (that is, exhibiting the least selectivity) is not the most 
reactive. Hexafluoro system 65, though forming 67 with a rate constant 1.5 times that of 
analogous closure of 68 to 70, does so only to an extent of 10.6% of total cyclized 
products. 

A combined ab initio I molecular mechanics approach has allowed for accurate 
regiochemical predictions for a number of alkyl and heteroalkyl intramolecular radical 
additions. 66 In order to examine the effect of the degree and location of fluorine 
substitution on transition structure geometry and energetics, as well as on activation 
barriers and reaction enthalpy, the "chair-like" and "boat-like" 5-exo and 6-endo 
cyclization transition structures for the parent hydrocarbon and various fluorinated 
5-hexenyl systems, along with their respective open-chain radicals and products of 
5- and 6-membered ring closure, have been investigated with ab initio techniques. 

In accordance with a UMINDO/3 investgation of Bischof, 62 the lowest energy 
conformation of 5-hexenyl radical 1 was found to be an all-rrans methylene chain in a 
gauche orientation with the internal vinyl hydrogen (Figure 4-16.) Alignment of the singly 
occupied orbital of 1 with the adjacent C-H bond was found to be slightly preferred (ca. 
0.1 kcal mol" 1 ) over similar C-C alignment at the UHF/6-31G(d) + ZPE level. 

From the calculated structures and energies of cyclization products 
(cyclopentylmethyl and cyclohexyl radicals) it was possible to compute energies of 
reaction for hydrocarbon 1 and its fluorinated analogues. Total energies of reactant and 



93 

product radicals and exothermicities of 5-exo and 6-endo cyclization for 1, 128, 152, and 
65 are provided below in Table 4-3. 





r(C=C) = 1.318A 



Figure 4-16. Lowest Energy Conformation of 5-Hexen-1-yl Radical 1; SOMO (On Right) 
Aligned with Adjacent C-H Bond. UHF/6-31G(d) Optimized Geometry. 



As expected from C-C BDE data, (Table 3-4) intermolecular additions of the 
fluorinated species are, as a whole, more exothermic than for parent 1. However, no 
direct correlation exists between either absolute rates of 5-exo and 6-endo addition or 
relative percentage of 6-membered ring formation and its corresponding reaction 
exothermicity. Although a steady increase in both k C5 and k C6 is observed along the 
series (1 -+ 128 -> 152 -> 65), both cyclizations of 65 are predicted to be less 
exothermic than those of 152. Furthermore, relative enthalpies (AE^s) - AEonft,«j) are 
found to rise with the degree of fluorination, favoring 6-endo closure in consistent 
manner for both levels of theory employed. This is at variance with the lesser extent of 
6-endo closure in 65 compared to 152. 

Total, zero-point, and relative energies along with pertinent geometrical 
parameters for the UHF/6-31G(d) cyclization transition structures of 1, (depicted in 
Figures 1-11 - 1-14) 128, 140, 152, and 65 are reported in Table 4-4. Although the 
calculated energy differences between "chair" and "boat" forms of either 5-exo or 6-endo 
transition structures are quite consistent among the theoretical methods, energies of the 
"6-endo-chair" and "6-endo-twist-boaf structures relative to the "5-exo-chair" and 
"5-exo-boar appear to be overestimated at the PMP2/6-311G(d,p)//UHF/6-31G(d) level 
compared to both UHF and QCISD(T) results. Bearing this in mind, relative transition 



94 



Table 4-3. Total and Zero-Point Energies and Energies of Reaction for 5-Exo and 
6-Endo Cyclizations of Hydrocarbon and Partially Fluorinated 5-Hexenyl Radicals. 



Radical 


F a 
£tg_t_ 


ZPE" 


AE™ (5-Exo) c 


AE™ (6-Endo) c 


AAErxn. 


1 


-233.543768 
(-234.461575) 
[-234.391386] 


0.161154 


-13.20 
(-18.58) 
[-15.90] 


-19.45 

(-23.44) 
[-21.22] 


6.25 
(4.86) 
[5.32] 


2 


-233.567930 
(-234.494315) 
[-234.419913] 


0.164664 








3 


-233.579940 
(-234.504105) 
[-234.431547] 


0.166947 








128 


-431.260147 
(-432.620036) 


0.148540 


-19.91 
(-23.09) 


-27.50 
(-29.51) 


7.59 
(6.42) 


137 


-431.291745 
(-432.656712) 


0.148402 








158 


-431.305809 
(-432.668897) 


0.150592 








152 


-628.970751 
(-630.769253) 


0.132048 


-20.29 
(-23.38) 


-29.15 
(-31.30) 


8.86 
(7.92) 


154 


-629.003100 
(-630.806518) 


0.132061 








156 


-629.018995 
(-630.820937) 


0.134061 








65 


-826.678248 
(-828.915986) 


0.115172 


-18.68 
(-21.77) 


-28.22 

(-30.36) 


9.54 
(8.59) 


66 


-826.708181 
(-828.950843) 


0.115350 








67 


-826.724889 
(-828.966024) 


0.117035 









3 In Hartrees. UHF/6-31G(d); PMP2/6-311G(d,p)//UHF/6-31G(d) Values in Parentheses, 

QCISD(T)/6-31G(d)//UHF/6-31G(d) Values in 

6-31 G(d) Vibrational Frequencies. 

6-311G(d,p)//UHF/6-31G(d) Values in 

6-31 (d) Values in Brackets. d (AE^ 5) 

6-311G(d,p)//UHF/6-31G(d) Values in 

6-31 (d) Values in Brackets. 



Brackets. In Hartrees, from UHF/ 
In kcal mor 1 ; UHF/6-31G(d); PMP2/ 
Parentheses, [QCISD(T)/6-31 1G(d,p)]7/UHF/ 
AErxno.6)) in kcal mol" 1 ; UHF/6-31G(d); PMP2/ 
Parentheses, [QCISD(T)/6-31 1G(d,p)]7/UHF/ 



95 



Table 4-4. UHF/6-31G(d) Geometric Parameters and Energies of Transition Structures 
for Hydrocarbon and Fluorocarbon 5-Hexenyl Cyclizations. 

Radical r (C-C) r (C=C) z C-C-C Jb^I ZPE AS E^ 

(A) (A) (Peg.) (au) (ay) (eu) 

1 a 2.186 1.393 109.7 -233.525546 0.161472 78.11 

(-234.455098) 
[-234.379299] 

1" 2.192 1.394 108.2 -233.522908 0.161317 78.91 1.57 

(-234.452664) (1.44) 

[-234.376772] [1.38] 

1 C 2.260 1.384 98.4 -233.521897 0.161801 76.91 2.24 

(-234.450180) (3.27) 

[-234.375531] [2.47] 

1 d 2.245 1.386 100.0 -233.517681 0.161736 77.77 5.08 

(-234.446462) (5.56) 

[-234.371537] [4.81] 

128 a 2.192 1.387 108.9 -431.244513 0.146381 86.27 

(-432.614627) 

128" 2.199 1.388 107.0 -431.241713 0.146261 86.97 1.69 

(-432.611948) (1.61) 

128 c 2.259 1.383 96.8 -431.241187 0.146926 85.37 2.39 

(-432.609955) (3.24) 

128 d 2.240 1.385 99.9 -431.237289 0.146800 85.89 4.76 

(-432.606471) (5.35) 

140 a 2.190 1.391 109.2 -431.248280 0.145212 85.54 

(-432.615856) 

140" 2.193 1.393 107.4 -431.245575 0.145080 86.64 1.62 

(-432.613199) (1.59) 

140 c 2.272 1.382 98.0 -431.247082 0.145752 83.98 1.05 

(-432.613844) (1.57) 

140" 2.257 1.384 99.6 -431.241993 0.145540 84.99 4.13 

(-432.608898) (4.55) 

152 a 2.193 1.386 108.6 -628.954014 0.129951 93.47 

(-630.762361) 

152" 2.195 1.387 106.5 -628.951440 0.129840 94.44 1.55 

(-630.759854) (1.51) 

152 c 2.269 1.381 96.3 -628.953289 0.130521 92.33 0.77 

(-630.760695) (1.36) 



96 
Table 4-4- continued 



Radical 


r (C-C) 
(A) 


r (C=C) 
(A] 


Z C-C-C 
(Deq.) 


F e 

£tot_ 

(au) 


ZPE 

{au) 


152" 


2.249 


1.383 


100.0 


-628.947730 
(-630.754984) 


0.130365 


65 a 


2.195 


1.387 


107.9 


-826.661679 
(-828.909230) 


0.113158 


65" 


2.199 


1.389 


106.3 


-826.659545 
(-828.907252) 


0.113017 


65 c 


2.271 


1.381 


96.8 


-826.660359 
(-828.906567) 


0.113677 


65 d 


2.252 


1.383 


99.3 


-826.656303 
(-828.902416) 


0.113613 



AS Erej_ 

(eu) 

78.91 4.17 
(4.86) 

76.91 



77.77 1.26 
(1.16) 

86.27 1.12 
(1.96) 

86.97 3.62 
(4.53) 



a 5-Exo-chair. b 5-Exo-boat. ° 6-Endo-chair. d 6-Endo-twist-boat. e UHF/6-31G(d); 
PMP2/6-311G(d,p)//UHF/6-31G(d) Values in Parentheses, QCISD(T)/6-31G(d)//UHF/ 
6-31G(d) Values in Brackets. f In kcal mol" 1 ; UHF/6-31G(d); PMP2/6-311G(d,p)// 
UHF/6-31G(d) Values in Parentheses, [QCISD(T)/6-311G(d,p)]7/UHF/6-31(d) Values in 
Brackets. 

state free energies were obtained for each structure based on the UHF/6-31G(d) 
energies and entropies, and regiochemical ratios determined based on a Boltzmann 
distribution including all four transition structures. 

In a similar manner, the above structures along with those of 1,1,2,2,3,3,4,4- 
octafluoro radical 68, perfluorinated system 71, and partially fluorinated 74 were also 
investigated with the UHF/4-31G model, this smaller basis set implemented due to the 
size of the latter three systems. Relative free energies for all systems investigated are 
provided in Table 4-5, with resultant predicted and experimental 5-exo : 6-endo ratios for 
each cyclization given in Table 4-6. 

Agreement between the computed and observed values is quite remarkable, 
even with the relatively small 4-31 G basis, the largest deviation from experiment 
approximately eight percent. From these results, in can be inferred that the 
regioselectivities for a variety of radical cyclizations (hydrocarbon, fluorocarbon, or 
otherwise) can be predicted with a reasonably high degree of confidence at minimal 



97 



Table 4-5. Relative Free Energies of Hydrocarbon and Fluorocarbon Transition 
Structures (kcal mol" 1 ) at 303 K. 



Radical 


5-Exo-chair 


5-Exo-boat 


6-Endo-chair 


6-Endo-twist-boat 


1 


0.0 
(0.0) 


1.22 
(1.33) 


3.10 
(2.84) 


5.33 
(5.19) 


128 


0.0 

(0.0) 


1.44 
(1.48) 


2.91 
(2.67) 


4.94 
(4.88) 


140 


0.0 
(0.0) 


1.23 
(1.29) 


1.35 

(1.53) 


4.28 
(4.30) 


152 


0.0 
(0.0) 


1.26 
(1.26) 


1.07 
(1.12) 


4.57 
(4.37) 


65 


0.0 
(0.0) 


1.15 
(1.22) 


1.68 
(1.45) 


4.02 
(3.96) 


68 


0.0 
(0.0) 


0.72 


0.83 


4.01 


71 


0.0 
(0.0) 


0.61 


2.95 


6.66 


74 


0.0 
(0.0) 


1.14 


2.10 


6.57 



Note: UHF/4-31G; UHF/6-31G(d) Values in Parentheses. 

computational expense and without the introduction of experimentally-based 
parameters. This is due in part to the fact that the degree of spin contamination in each 
of these structures, though rather high (with <S A 2> values ranging in magnitude from 
approximately 1.0 to 1.1) is fairly consistent. This is not true, however, of the ground 
state starting structures, which adversely affects computed activation barriers as 
discussed below. 

The experimental activation barriers for 5-exo and 6-endo closure of 1 are 6.8 
and 8.5 kcal mol" 1 , respectively. 6364 Calculated energies of activation for the parent 
hydrocarbon and number of fluorinated analogues are found in Table 4-7. 

It can be seen from the results of 1 that absolute barriers for radical cyclization 
reactions are extremely difficult to model theoretically. Even with the inclusion of 



98 



Table 4-6. Computed and Experimental Regiochemical Ratios at 303 K for Hydrocarbon 
and Fluorocarbon 5-Hexenyl Radical Cyclizations. 



Radical 



5-Exo : 6-Endo Ratio 





Predicted 


1 


100:0 
(99.3 : 0.7) 


128 


100:0 
(98.9: 1.1) 


140 


91.3:8.7 
(93.3 : 6.7) 


152 


86.9: 13.1 
(87.8: 12.2) 


65 


94.8 : 5.2 
(92.6 : 7.4) 


68 


83.8 : 16.2 


71 


99.5 : 0.5 


74 


97.4 : 2.6 



Observed 
98:2 

> 96 : < 4 
90.9 : 9.1 

82.1 : 17.9 
89.4: 10.6 

75.9:24.1 

> 96 : < 4 

> 96 : < 4 
3 UHF/4-31G; UHF/6-31G(d) Values in Parentheses. 

electron correlation, accuracy leaves quite a bit to be desired. These results are in 
accord with the those of Wong and Radom, 41 whose systematic investigation of 
intermolecular additions of small model radicals to olefins has demonstrated that high 
levels of theory are required to accurately reproduce (and predict) activation barriers for 
radical reactions. This is due in part to the differing degree of spin contamination in the 
starting radicals and transition structures as mentioned previously, the former 
possessing "acceptable" <S A 2> values in the range of 0.76 - 0.79, versus that of 0.75 for 
a "pure" doublet radical. Such differential spin contamination has a disastrous effect on 
activation barriers computed using perturbation theory. For example, UMP2/6- 
311G(d,p)//UHF/6-31G(d) yields values of 13.5 and 17.3 kcal mol" 1 , respectively, for 1; 
calculations at the UMP4SDQ/6-311G(d,p) level on UHF/6-31G(d) geometries fare only 
slightly better with values of 12.1 and 15.3, only a modest improvement over UMP2/ 



99 



Table 4-7. Calculated Activation Barriers for 5-Exo and 6-Endo Cyclizations of 
Hydrocarbon and Fluorocarbon 5-Hexen-1-yl Radicals. 

Radical UHF/4-31G UHF/6-31G(d) PMP2/6-31 1 G(d,p)//UHF/6-31 G(d) 

1 5-exo : 10.67 5-exo : 1 1 .61 5-exo : 4.24 

6-endo : 13.38 6-endo : 14.09 6-endo : 7.51 

128 5-exo : 6.10 5-exo : 8.60 5-exo : 2.18 

6-endo: 8.75 6-endo: 10.99 6-endo : 5.42 

140 5-exo : 6.16 5-exo : 9.33 5-exo : 3.15 

6-endo: 6.89 6-endo: 10.10 6-endo : 4.51 

152 5-exo : 6.56 5-exo : 9.27 5-exo : 3.1 1 

6-endo : 7.93 6-endo : 10.39 6-endo : 5.07 

65 5-exo : 8.86 

6-endo : 9.41 

Note: In kcal mol" 1 . Includes 0.8929 ZPE Correction. 



6-31 1G(d,p) and still worse than the Hartree-Fock results. The slow convergence of the 
UMPn series in systems suffering from severe spin contamination has been noted by 
Nobesefa/. 196 

The spin-projected PMP2/6-311G(d,p) values, though faring somewhat better in 
the case of 1, do not reproduce the correct trends for either 5-exo or 6-endo reactivity 
along the fluorinated series. Methods of high accuracy such as coupled cluster (CC) 
and quadratic configuration interaction (QCI) have also been shown to be robust against 
spin contamination 197 198 but at significant (and, for larger systems, prohibitive) expense. 
Results of QCISD(T) barrier calculations for 1 and 128 are provided in Table 4-8, the 
former requiring 7-10 and the latter 28-35 hours of CPU time and approximately 2-4 GB 
of disk space on the Cray C90 at the San Diego Supercomputer Center for each 
calculation of the energies of starting species and transition structures. As expected, 
however, the agreement between theory and experiment for 1 is much improved at this 
level, especially with partial inclusion of basis extension effects by evaluation at the 
PMP2/6-311G(d,p) level. 



100 



Table 4-8. Activation Barriers for 5-Exo and 6-Endo Cyclization of 1 and 128 at the 
QCISD(T) Level. 

Radical QCISD(T)/6-31G(d) fQCISDm/6-311Gfd.p)T a 

1 5-exo : 7.76 5-exo : 7.30 

6-endo : 10.31 6-endo : 9.77 

128 5-exo: 4.98 5-exo : 5.21 

6-endo : 7.21 6-endo : 7.75 

Note: UHF/6-31G(d) Geometries, kcal mol" 1 . Includes 0.8929 ZPE Correction. 

a E[QCISD(T)/6-311G(d,p)]' * E[QCISD(T)/6-311G(d,p)] = E[QCISD(T)/6-31G(d)] + 

E(PMP2/6-311G(d,p)) - E(PMP2/6-31G(d)). 



With such wide variations in variations in regiochemistry observed in the 
cyclizations of 1, it is difficult to provide a rationale which satisfactorily accounts for all of 
the data. This is especially so given the insensitivity of cyclization transition structure 
geometry to substitution (Table 4-4 demonstrates the similarity in geometric trends to 
those of intermolecular additions). However, the predictive value of the ab initio results 
is encouraging, allowing for the potential design of precursors with desired regiochemical 
behavior in cyclization and cyclopolymerization reactions. 

Conclusion 

Based on absolute rate constants of hydrogen abstraction from tributylstannane 
and rr/s(trimethylsilyl)silane by partially fluorinated alkyl radicals, absolute rate constants 
for 5-exo and 6-endo intramolecular addition reactions of partially fluorinated 5-hexen-1- 
yl radicals have been obtained through competitive kinetic methods. Regardless of 
either the degree or location of fluorine substitution on the 5-hexenyl chain, closure 
proceeds predominantly in an exo fashion, although in some cases substantial 
competing 6-endo cyclization is observed. 

The kinetics of cyclization are found to follow those observed in the bimolecular 
addition of partially fluorinated radicals to alkenes. The pyramidal nature of a,a- 
difluoroalkyl radicals in combination with increased electrophilic character as function of 



101 

incremental fluorine substitution are responsible for their enhanced reactivity over 
hydrocarbon analogues. 

Consistent with high level computational studies on model radical addition 
reactions, it is found that absolute activation barriers for intramolecular addition reactions 
of 5-hexenyl radicals prove extremely difficult to model theoretically. Predictions of 
cyclization regiochemistry based on relative ab initio transition state free energies may 
be made with a reliable degree of accuracy. 



CHAPTER 5 



EXPERIMENTAL 
General Methods- Experimental 

NMR nuclear magnetic resonance (NMR) spectra (300 MHz, 75 MHz and 282 
MHz for 1 H, 13 C, and 19 F respectively) are reported in parts per million ppm downfield (8) 
versus tetramethylsilane (TMS) for 1 H and 13 C, and in ppm upfield (<|>) versus CFCI 3 for 
19 F. All NMR spectra were recorded on Varian VXR-300 or Gemini-300 spectrometers. 

Preparative gas chromatographic (GC) separations were carried out with a 20 
foot x 0.25 inch copper column packed with 20% SE-30 on Chromosorb P and 
performed on a Varian Aerograph A-90 gas chromatograph equipped with a thermal 
conductivity detector. 

High resolution mass spectra were obtained on a Finnegan MAT-95 
spectrometer. 

Ultraviolet (UV) spectra were obtained on a Perkin-Elmer Lambda 9 UVA/IS/NIR 
spectrophotometer. 

All reagents, unless otherwise specified, were purchased from Aldrich, Fisher, 
PCR, or Acros, and used as received. Styrene (Fisher) was freed from inhibitor by 
passage through a column of neutral alumina. Dichloromethane was distilled from 
calcium hydride and used immediately. Diethyl ether and tetrahydrofuran (THF) were 
distilled from sodium benzophenone ketyl and used immediately. Benzene was distilled 
from lithium aluminum hydride and stored over 4 A molecular sieves. Chloroform, 
dimethylsulfoxide, and tetraglyme were commercial anhydrous grade. All reactions were 
performed under an inert atmosphere of argon. 



102 



103 
General Methods- Theoretical 

All ab initio calculations were performed with the Gaussian92 and Gaussian94 
program systems 199 on Cray Y-MP 4/32 or Cray C90 supercomputers, an IBM RS/6000 
SP2 cluster, or Intel-based PCs. Geometry optimizations were performed using 
standard gradient techniques. All stationary points were characterized by harmonic 
frequency analysis, minima and transition structures giving rise to zero or exactly one 
negative eigenvalue, respectively, in the second derivative matrix. All post-Hartree-Fock 
calculations were performed with core orbitals frozen. Hartree-Fock frequencies and 
resultant zero-point energies have been scaled by a factor of 0.8929. 

Density functional theory calculations were performed with the Gaussian94 
program on an IBM RS/6000 SP2 cluster or Intel-based PCs, implementing the 
(75,302)p pruned integration grid. Zero-point energy corrections were scaled by 0.9806 
as suggested by Scott and Radom and by Bauschlicher and Partridge. 200201 For bond 
dissociation energy calculations, a thermal correction of ART (for C-C BDEs) or 2.5RT 
(for C-H BDEs) has been applied as recommended by Hehre et al 202 Where applicable, 
reported BDEs correspond to those resulting from the lowest electronic energy 
conformer of the closed-shell species and / or radical. 

Synthetic Procedures 

Preparation of 1,3-Dibromo-1,1-difluorohexane (79) 

A Carius tube of approximately 200 mL capacity equipped with a small magnetic 
stir bar was charged with 0.14 g (1.42 x 10" 3 mol) cuprous chloride, 4.36 g (7.12 x 10" 2 
mol) ethanolamine, 12 mL tert-butanol, 10.0 g (1.42 x 10" 1 mol) 1-pentene, and 59.84 g 
(2.85 x 10~ 1 mol) dibromodifluoromethane. The tube was flushed with nitrogen and 
flame-sealed; upon swirling, a deep blue coloration was observed. The tube was 
immersed halfway into a silicon oil bath preheated at 85° C and allowed to stir for 48 
hours, during which time the coloration turned from deep blue to olive green to brown. 



104 

(Caution: this procedure should be performed behind a safety shield). The tube was 
cooled in an ice bath, opened, and the contents transferred to a 250-mL Erlenmeyer 
flask (at this point, unreacted dibromodifluoromethane may be recovered by distillation) 
and the tube rinsed with three 50 mL portions of hexanes. All organic material (which 
consisted of a cloudy yellow-green supernatant and a brown resin) was filtered through 
50 mL of silica gel, which was rinsed with two additional 50 mL portions of hexanes. The 
resulting colorless filtrate was concentrated by rotary evaporation and subject to reduced 
pressure fractional distillation through a 15 cm Vigreux column. A total of 22.79 g 
(57.3%) 79 was obtained as a colorless liquid, bp 80-85° C / 25 mm Hg. 

1,3-Dibromo-1,1-difluorohexane (79): 1 H NMR: 5 0.96 (3H, t, 3 J HH = 7.42 Hz), 
1.50 (2H, m), 1.86 (2H, m), 3.01 (2H, m), 4.24 (1H, m); 13 C NMR: 8 13.2, 20.4, 40.5, 
46.6, 52.7 (t, 2 J CF = 21.8 Hz), 120.6 (t, 1 J CF = 306.9 Hz); 19 F NMR: f -43.2 (m); HRMS for 
C 6 H 10 F 2 Br 2 : calc. 277.9117, calc. (M-Br) 198.9934, found 198.9983; CHN for C 6 H 10 F 2 Br 2 : 
calc. 25.74% C, 3.60% H, found 25.53% C, 3.43% H. 

Preparation of 1-Bromo-1,1-difluorohexane (80) and 1,1-Difluorohexane (81) 

A 250 mL three-necked round-bottomed flask equipped with an ice-water 
condenser, argon inlet, and magnetic stir bar was charged with 20.0 g (7.14 x 10' 2 mol) 
1,3-dibromo-1,1-difluorohexane (79) dissolved in 100 mL anhydrous dimethylsulfoxide. 
A total of 10.8 g (2.85 x 10" 1 mol) of sodium borohydride was then added in small 
portions with vigorous stirring over the course of 1 hour, during which time the flask 
became warm and a semisolid gel was observed to form. After the addition was 
complete, the bath temperature was raised to 70° C over the course of 1 hour and 
heating continued for an additional 6 hours ( 19 F NMR analysis of a small aliquot of the 
reaction mixture at this time showed complete consumption of starting material.) The 
flask was cooled to room temperature, the contents transferred to a 1 L Erlenmeyer 
flask, and the reaction quenched with chips of ice. The resulting mixture was carefully 



105 

acidified with concentrated aqueous HCI, 100 mL ether was added, and the aqueous / 
DMSO layer extracted with three 50 mL portions of ether. The combined ether layers 
were washed with three 25 mL portions of water, dried over MgS0 4 , and subject to 
ambient pressure fractional distillation through a 15 cm Vigreux column. After 
concentration in this way, 8.91 g (62.1%) 80 (contaminated with a small amount of 81) 
was obtained as a colorless liquid, bp 125-128° C. Preparative GC separation afforded 
analytically pure samples of each. 

1-Bromo-1,1-difluorohexane (80): 1 H NMR: 8 0.92 (3H, t, 3 J HH = 7.5 Hz), 1.35 
(4H, m), 1.62 (2H, m), 2.33 (2H, m); 13 C NMR: 8 13.8, 22.3, 23.6, 30.6, 44.3 (t, 2 J CF = 
21.1 Hz), 123.3 (t, 1 J CF = 303.5 Hz); 19 F NMR: f -43.9 (t, 3 J FH = 14.7 Hz); HRMS for 
C 6 HnF 2 Br: calc. 200.0012, calc. (M-Br) 121.0829, found 121.0832; CHN for C 6 H 11 F 2 Br: 
calc. 35.84% C, 5.51% H, found 35.47% C, 5.54% H. 

1.1-Difluorohexane(81): 1 H NMR: 5 0.91 (3H, t, 3 J HH = 6.9 Hz), 1 .34 (4H, m), 1 .45 
(2H, m), 1 .80 (2H, m), 5.79 (1 H, tt, 3 J HH = 4.5 Hz, 2 J HF = 57 Hz); 13 C NMR: 8 1 3.8, 21 .8 (t, 
3 J CF = 5.5 Hz), 22.4, 31.2, 34.1 (t, 2 J CF = 20.6 Hz), 117.5 (t, 1 J CF = 237.4 Hz); 19 F NMR: <)> 
-116.3 (dt, 3 J FH = 17.1 Hz, 2 J FH = 56.2 Hz); HRMS for C 6 H 12 F 2 : calc. 122.0907, calc. 
(M-H) 121.0829, found 121.0825. 

Preparation of 1-Phenyloctan-3-ol (86) 

Into a flame-dried 300 mL three-necked round-bottomed flask equipped with a 
125 mL pressure-equalizing addition funnel, condenser with argon inlet, and magnetic 
stir bar was dispensed 3.94 g (1.62 x 10" 1 mol) Mg turnings and a small crystal of iodine. 
(2-Bromoethyl)benzene (10.0 g, 5.40 x 10" 2 mol) was dissolved in 100 mL of anhydrous 
ether and added to the funnel. After addition of a small amount of solution and initiation 
of the reaction, addition was continued at a rate that maintained gentle reflux, after which 
time the mixure was refluxed for an additional 2 hours. The flask was then cooled to 
room temperature and 5.41 g (5.40 x 10~ 2 mol) of hexanal was dissolved in 50 mL 



106 

anhydrous ether, added to the addition funnel and dispensed dropwise at a rate that 
maintained gentle reflux. The mixture was refluxed for an additional 2 hours, cooled to 
room temperature, and quenched with a mixture of 100 g of ice and 50 mL concentrated 
aqueous HCI. The layers were separated, the aqueous layer extracted with three 50 mL 
portions of ether, and the combined organic extracts washed twice with 20 mL 5% 
aqueous NaHC0 3 and once with 20 mL of water. The organic layer was dried over 
MgS0 4 , the solvent rotary evaporated, and the resulting liquid subjected to reduced 
pressure fractional distillation through a 15 cm Vigreux column. 8.75 g (78.7%) 86 was 
obtained as a colorless liquid, bp 148-151° C / 0.1 mm Hg. 

1-Phenvloctan-3-ol (86): 1 H NMR: 8 0.94 (3H, t, 3 J HH = 5.7 Hz), 1.34 (4H, m), 1.51 
(4H, m), 1.81 (2H, m), 2.76 (2H, m), 3.67 (1H, m), 7.28 (5H, m); 13 C NMR: 8 14.0, 22.6, 
25.2, 31.8, 32.0, 37.5, 39.0, 71.4, 125.7, 128.3, 128.4, 142.2; HRMS for C 14 H 22 0: calc. 
206.1671, found 206.1672. 

Preparation of 1-Phenyloctan-3-one (87) 

Into a 100 mL round-bottomed flask was placed 3.8 g (1.84 x 10" 2 mol) of 
1-phenyloctan-3-ol (86) dissolved in 10 mL ether. Jones' reagent (8.5 mL, previously 
prepared with of 5.0 g Na 2 Cr 2 7 and 3.65 mL concentrated H 2 S0 4 diluted with water to a 
total volume of 25 mL) was added dropwise to the alcohol solution with stirring, and 
allowed to react for 4 hours at room temperature. The dark green reaction mixture was 
diluted with 20 mL ether and 20 mL water. The aqueous layer was extracted with three 
20 mL portions of ether, and the combined organic extracts washed twice with 20 mL 
saturated NaHC0 3 , once with 20 mL brine, and dried over MgS0 4 . Rotary evaporation 
of the solvent followed by reduced pressure fractional distillation through a 10 cm 
Vigreux column afforded 3.34 g (89.0%) 87, bp 108-111° C / 0.35 mm Hg. 

1-Phenvloctan-3-one (87): 1 H NMR: 8 0.87 (3H, t, 3 J HH = 6.9 Hz), 1.26 (4H, m), 
1.55 (2H, overlapping tt, 3 J HH = 7.5 Hz), 2.37 (2H, t, 3 J HH = 7.2 Hz), 2.72 (2H, m), 2.89 



107 

(2H, m), 7.23 (5H, m); 13 C NMR: 5 13.9, 22.4, 23.5, 29.8, 31.4, 43.0, 44.2, 126.0, 128.3, 
128.4, 141.2, 210.3; HRMS for C 14 H 20 O: calc. 204.1514, found 204.1532; CHN for 
C 14 H 20 O: calc. 82.30% C, 9.87% H, found 82.53% C, 10.14% H. 

Preparation of 3,3-Difluoro-1-phenyloctane (83) 

Into a 100 mL three-necked round-bottomed flask equipped with stir bar, rubber 
septum, and condenser with argon inlet was placed 2.0 g (9.80 x 10 3 mol) of 1- 
phenyloctan-3-one (87) dissolved in 15 mL anhydrous CH 2 CI 2 . Into the reaction mixture 
was slowly injected 1.74 g (1.08 x 10" 2 mol) diethylaminosulfurtrifluoride (DAST) with 
stirring, during which time the flask became slightly warm. The mixture was heated at 
reflux for 72 hours, over which time a dark amber coloration was observed. After 
cooling, the reaction mixture was carefully decanted onto 50 mL of ice water and diluted 
with 10 mL CH 2 CI 2 . The organic layer was separated and the aqueous layer extracted 
with three 20 mL portions of CH 2 CI 2 . The combined organic extracts were washed once 
with 10 mL of 10% aqueous NaHC0 3 and once with 10 mL water, dried over MgS0 4 , 
and the solvent rotary evaporated. Fractional reduced pressure distillation of the 
resulting liquid afforded 1.66 g (68.0%) 83, bp 100-105° C / 0.2 mm Hg. 

3,3-Difluoro-1-phenvloctane (83): 1 H NMR: 8 0.92 (3H, t, 3 J HH = 6.3 Hz), 1.33 
(4H, m), 1.48 (2H, m), 1.87 (2H, m), 2.13 (2H, m), 2.82 (2H, m), 7.27 (5H, m); 13 C NMR. 
5 13.9, 22.0, 22.4, 28.5, 31.5, 36.5 (t, 2 J CF = 25.1 Hz), 38.2 (t, 2 J CF = 25.5 Hz), 124.8 (t, 
1 J CF = 239.4 Hz), 126.1, 128.3, 128.5, 140.8); 19 F NMR: <|> -99.1 (overlapping tt, 3 J FH = 
17.1 Hz); HRMS for C 14 H 20 F 2 : calc. 226.1533, found 226.1529; CHN for C 14 H 20 F 2 : calc. 
74.30% C, 8.91% H, found 74.14% C, 9.15% H. 

Preparation of 1-Bromohexan-2-one (90) 

Into a 250 mL three-necked round-bottomed flask equipped with an argon inlet, 
magnetic stir bar, and 100 mL pressure-equalizing addition funnel was placed 10.0 g 
(9.98 x 10" 2 mol) of 2-hexanone, 40 mL glacial acetic acid, and 9.75 g (1.62 x 10" 1 mol) 



108 

urea. The flask was cooled to 0° C and 17.15 g (1.07 x 10 1 mol) of bromine was 
introduced dropwise into the reaction mixture over the course of 1 hour. After the 
addition was complete, the cooling bath was removed and the reaction mixture allowed 
to warm to room temperature and stirred overnight. The contents were then transferred 
to a separatory funnel, diluted with 200 mL water, and extracted with four 50 mL portions 
of CH 2 CI 2 . The combined organic extracts were washed with three 20 mL portions of 
10% aqueous NaHC0 3 , dried over MgS0 4 , and the solvent rotary evaporated. 
Fractional reduced pressure distillation through a 15 cm Vigreux column yielded 12.91 g 
(72.2%) 90, contaminated with ca. 6% (by 1 H NMR) of the 3-bromo isomer, bp 85-88° C / 
15 mm Hg. Column chromatography (silica gel, 20% ethyl acetate in hexanes) of a 
small sample yielded pure 90, which was used in characterization. 

1-Bromohexan-2-one (90) : 1 H NMR: 5 0.91 (3H, t, 3 J HH = 7.5 Hz), 1.33 (2H, 
overlapping tt, 3 J HH = 7.5 Hz), 1.60 (2H, overlapping tt, 3 J HH = 7.5 Hz), 2.64 (2H, t, 3 J HH = 
7.2 Hz), 3.88 (2H, s); 13 C NMR: 5 13.7, 22.1, 25.9, 34.3, 39.5, 202.2; HRMS for 
C 6 HnBrO: calc. 177.9993, found 177.9934; CHN for CeHnBrO: calc. 40.25% C, 6.19% 
H, found 40.10% C, 6.19% H. (Caution: a-bromoketones are powerful lachrymators). 

Preparation of 1-Bromo-2,2-difluorohexane (91) 

A 100 mL three-necked round-bottomed flask equipped with magnetic stir bar, 
condenser with argon inlet, and rubber septum was charged with 5.0 g (2.79 x 10" 2 mol) 
1-bromohexan-2-one (90) dissolved in 20 mL anhydrous CHCI 3 . DAST (9.15 g, 
5.68 x 10" 2 mol) was slowly injected into the reaction mixture, which was then heated at 
50° C for 48 hours. The flask was cooled and the contents carefully dispensed into 50 
mL of ice water. The layers were separated, the aqueous layer extracted with three 5 
mL portions of CH 2 CI 2 , and the combined organic extracts washed twice with 10 mL 
portions of 10% aqueous NaHC0 3 and once with 10 mL of water. Drying over MgS0 4 , 
evaporation of the halogenated solvents by ambient-pressure distillation through a 10 






109 

cm Vigreux column, and fractional, ambient pressure distillation afforded 3.35 g (59.7%) 
91 as a colorless liquid, bp 120-122° C. 

1-Bromo-2.2-difluorohexane (91) : 1 H NMR: 5 0.94 (3H, t, 3 J HH = 7.2 Hz), 1.42 
(4H, m), 2.02 (2H, m), 3.51 (2H, t, 3 J HF = 13.2 Hz); 13 C NMR: 13.7, 22.3, 24.1, 31.4 (t, 
2 J CF = 34.1 Hz), 34.2 (t, 2 J CF = 24.0 Hz), 121.5 (t, 1 J CF = 240.9 Hz); 19 F NMR: + -99.2 (m); 
HRMS for CeHnFzBr: calc. 200.0012, calc. (M-H) 198.9934, found 198.9974; CHN for 
CeHnFzBr: calc. 35.84% C, 5.51% H, found 36.00% C, 5.48% H. 

Preparation of 1-lodo-2,2-difluorohexane (92) 

1-Bromo-2,2-difluorohexane (91) (2.0 g, 9.95 x 10" 3 mol) was placed into a thick- 
walled Carius tube of approximately 150 mL capacity, along with a small magnetic stir 
bar. 100 mL of a hot saturated solution of sodium iodide in acetone was added to the 
tube, which was then cooled in a dry ice-isopropanol slush and flame-sealed. After 
warming to room temperature, the tube was immersed halfway in an oil bath atop a 
magnetic stirrer. The mixture was stirred at 85-90° C for 96 hours (a safety shield is 
recommended) at which time the tube was cooled, opened, and the contents transferred 
to a 250 mL round-bottomed flask. Most of the solvent was removed by rotary 
evaporation, and the remaining residue taken up in a mixture of 50 mL of water and 50 
mL of ether. The layers were separated, the aqueous layer extracted with three 50 mL 
portions of ether, and the combined organic extracts washed with 20 mL of water. 
Drying over MgS0 4 , rotary evaporation of the solvent, and fractional reduced-pressure 
distillation afforded 2.16 g (87.5%) of very pure 92, bp 101-102° C / 68 mm Hg. 

1-lodo-2.2-difluorohexane (92V 1 H NMR: 5 0.93 (3H, t, 3 J HH = 7.2 Hz); 1.41 (4H, 
m), 2.06 (2H, m), 3.40 (2H, t, 3 J HF = 14.4 Hz); 13 C NMR: 5 3.95 (t, 2 J CF = 31.5 Hz), 13.7, 
22.3, 24.4, 35.0 (t, 2 J CF = 24.5 Hz), 121.1 (t, 1 J CF = 240.8 Hz); 19 F NMR: <j> -94.9 
(overlapping tt, 3 J FH = 17.1 Hz); HRMS for CjHiiF^: calc. 247.9833, found 247.9862; 
CHN for CeHnFjl: calc. 29.05% C, 4.47% H, found 28.82% C, 4.48% H. 



110 

Preparation of 2,2-Difluorohexane (93) 

1-Bromo-2,2-difluorohexane (91) (1.0 g, 4.97 x 10" 3 mol) was dissolved in 5 mL 
benzene in a 25 mL round-bottomed flask equipped with septum-capped sidearm inlet 
and magnetic stir bar, and attached to a small distillation apparatus equipped with with 
ice water condenser and fractionating column. The bath temperature was raised to 
60° C and 1.60 g (5.50 x 10 3 mol) nBu 3 SnH and 0.01 g (6.09 x 10" 5 mol) 2,2'- 
azobisisobutyronitrile (AIBN) dissolved in 0.5 mL benzene was added slowly via syringe 
through the rubber septum. When the addition was complete, the mixture was stirred for 
an additional 30 minutes and the bath temperature quickly raised to 150° C. All volatile 
material was flash distilled and subjected to preparative GC separation which afforded 
analytically pure 93 as a colorless liquid. 

2,2-Difluorohexane (93) : 1 H NMR: 8 0.92 (3H, t, 3 J HH = 7.2 Hz), 1 .42 (4H, m), 1 .58 
(3H, t, 3 J HF = 18.3 Hz), 1.83 (2H, m); 13 C NMR: 8 13.8, 22.5, 23.2 (t, 2 J CF = 28.1 Hz), 24.9 
(t, 3 J CF = 4.5 Hz), 37.7 (t, 2 J CF = 25.1 Hz), 124.5 (t, 1 J CF = 236.4 Hz); 19 F NMR: <j> -90.9 
(m); HRMS for C 6 H 12 F 2 : calc. 122.0907, calc. (M-HF) 102.0845, found 102.0822. 

Preparation of 1-Phenyloctan-4-ol (96) 

In a manner similar to that of the preparation of 86, 10.0 g (5.02 x 10" 2 mol) of 
1-bromo-3-phenylpropane dissolved in 100 mL anhydrous ether was added to 3.67 g 
(1.51 x 10~ 1 mol) Mg turnings to which a crystal of iodine had been added. Subsequent 
addition of 4.54 g (5.27 x 10~ 2 mol) of valeraldehyde dissolved in 50 mL anhydrous ether 
followed by workup in the usual way afforded 7.36 g (71.1%) 96, bp 102-103° C / 0.05 
mm Hg. 

1-Phenvloctan-4-ol (96): 1 H NMR: 8 0.91 (3H, t, 3 J HH = 6.6 Hz), 1.52 (10H, 
overlapping m), 2.65 (2H, t, 3 J HH = 7.5 Hz), 3.62 (1H, m), 7.24 (5H, m); 13 C NMR: 14.0, 
22.7, 27.4, 27.8, 35.9, 37.0, 37.1, 71.7, 125.7, 128.2, 128.3, 142.4; HRMS for C 14 H 22 0: 



111 

calc. 206.1671, found 206.1671; CHN for C 14 H 22 0: calc. 81.50% C, 10.75% H, found 
81.72% C, 10.92% H. 

Preparation of 1-Phenyloctan-4-one (97) 

In a manner similar to that of the preparation of 87, 15 mL of Jones' reagent was 
added to 5.0 g (2.42 x 10~ 2 mol) 1-phenyloctan-4-ol (96) in 25 mL ether. Workup in the 
usual way followed by distillation at reduced pressure yielded 4.14 g (83.7%) 97, bp 123- 
125°C/0.4mmHg. 

1-Phenvloctan-4-one (97) : 1 H NMR: 5 0.96 (3H, t, 3 J HH = 7.2 Hz), 1.36 (2H, m), 
1.60 (2H, m), 1.97 (2H, overlapping tt, 3 J HH = 7.5 Hz), 2.46 (4H, m), 2.69 (2H, t, 3 J HH = 
7.2 Hz), 7.30 (5H, m); 13 C NMR: 13.8, 22.3, 25.2, 25.8, 35.0, 41.8, 42.5, 125.8, 128.3, 
128.4, 141.6, 211.0; HRMS for C 14 H 20 O: calc. 204.1514, found 204.1576; CHN C 14 H 20 O: 
calc. 82.30% C, 9.87% H, found 82.40% C, 9.83% H. 

Preparation of 4,4-Difluoro-1-phenyloctane (98) 

In a manner similar to the preparation of 83, 2.5 g (1.22 x 10" 2 mol) 
1-phenyloctan-4-one (97) and 3.93 g (2.44 x 10" 2 mol) DAST in 25 mL CH 2 CI 2 were 
refluxed for 48 hours. Workup in the usual way afforded 1.83 g (66.3%) 98, bp 
85-88° C/ 0.1 mmHg. 

4,4-Difluoro-1 -phenyloctane (98) : 1 H NMR: 8 0.89 (3H, t, 3 J HH = 6.9 Hz), 1.34 (4H, 
m), 1.80 (6H, m), 2.63 (2H, t, 3 J HH = 6.9 Hz), 7.22 (5H, m); 13 C NMR: 8 13.8, 22.5, 24.0 
(t, 3 J CF = 4.5 Hz), 24.38 (t, 3 J CF = 4.7 Hz), 35.4, 35.7 (t, 2 J CF = 26.0 Hz), 36.1 (t, 2 J CF = 
25.1 Hz), 125.2 (t, 1 J CF = 238.9 Hz), 125.9, 128.4 (2C, overlapping), 141.5; 19 F NMR: * - 
98.3 (overlapping tt, 3 J FH = 17.1 Hz); HRMS for C 14 H 20 F 2 : calc. 226.1533, found 
226.1533; CHN for C 14 H 20 F 2 : calc. 74.30% C, 8.91% H, found 74.40%C, 8.85% H. 



112 

Preparation of 1-Bromo-1,1,2,2-tetrafluorohexane (102) 

Into a 250-mL three-necked round-bottomed flask equipped with a magnetic stir 
bar, rubber septum, and condenser equipped with an argon inlet was added 36 mL 
tetraglyme and 30.0 g (1.28 x 10" 1 mol) of 6-bromo-5,5,6,6-tetrafluorohex-1-ene (101). A 
2.0 M solution of borane • dimethyl sulfide in diethyl ether (24 mL, 4.80 x 10' 2 mol) was 
slowly injected into the flask through the septum. Some bubbling was evident and the 
flask became slightly warm. This mixture was stirred for two hours at room temperature 
then heated at reflux with stirring overnight. The flask was then cooled to room 
temperature and 64 mL (59.3 g, 5.11 x Iff 1 mol) of hexanoic acid was slowly injected 
into the reaction mixture with stirring. Vigorous bubbling was evident and the flask 
became warm. Stirring was continued for 2 hours at room temperature then at reflux 
overnight. The mixture was distilled through a 15 cm Vigreux column at ambient 
pressure until the head temperature reached 140° C, at which time distillation ceased. 
The distillate was diluted with 50 mL of ether, washed with two 10 mL portions of 
saturated aqueous sodium bicarbonate and two 10 mL portions of water, dried, and the 
resulting solution distilled at ambient pressure. After removal of ether and residual 
dimethyl sulfide in this way, 24.16 g (79.6%) 102 was obtained as a colorless liquid 
boiling at 122-124° C. 

1 -Bromo-1 . 1 ,2,2-tetrafluorohexane (1 02): 1 H NMR: 8 0.95 (3H, t, 3 J HH = 7.2 Hz), 
1.41 (2H, m), 1.59 (2H, m), 2.07 (2H, m); 13 C NMR: 8 13.7, 22.3, 22.6, 30.1 (t, 2 J CF = 
22.5 Hz), 117.5 (tt, 2 J CF = 31.1 Hz, 1 J CF = 251.8 Hz,), 118.0 (tt, 2 J CF = 39.6 Hz, 1 J CF = 
309.5 Hz,); 19 F NMR: ((, -65.9 (2F, s), -112.6 (2F, t, 3 J FH = 19.5 Hz); HRMS for C 6 H 9 F 4 Br: 
calc. 235.9823, calc. (M+H) 236.9902, found 236.9738; CHN for C 6 H 9 F 4 Br: calc. 30.40% 
C, 3.83% H, found 30.72% C, 3.46% H. 



113 

Attempted Transhalogenation Reactions of 102 via Lithium-Halogen Exchange and 
Formation of 1,1,2-Trifluoro-1-hexene (103) 

In each instance, 1.0 g (4.22 x 10" 3 mol) 102 was dissolved in 25 ml_ dry diethyl 
ether and cooled in a bath of liquid nitrogen / pentane. A slight excess (1.1 equivalents; 
2.2 in the case of f-butyllithium) of the desired alkyllithium as a solution in diethyl ether 
was added very slowly to the reaction mixture via syringe. An solution of excess iodine 
in diethyl ether was then added dropwise to the mixture and allowed to warm to room 
temperature and stir overnight in each case. 19 F NMR analysis showed in all cases 
quantitative formation of 103. Use of carbon dioxide as an alternative electrophile was 
not successful; p-fluoride elimination of the lithiated species was too rapid, even at -100° 
C, to be successfully trapped. 

1.1.2-Trifluoro-1-hexene (103); 19 F NMR (CDCI 3 ): 8 -106.9 (1F, dd, 2 J FF = 90 Hz, 
3 J FF = 32 Hz), -125.9 (1F, dd, 2 J FF = 90 Hz, 3 J FF = 112 Hz), -174 .8 (1F, dtd, 3 J FF = 112 
Hz, 3 J FF = 32 Hz, 3 J FH = 22 Hz). 

Attempted lodofluorination of 103. 

To a solution of 1,1,2-trifluoro-1-hexene (103) (containing 9.04 x 10" 3 mol as 
judged by a 19 F NMR standard of hexafluorobenzene) in 10 mL of diethyl ether at 0° C 
was added 3.7 mL (3.65 g, 2.27 x 10" 2 mol) of triethylamine trihydrofluoride. N- 
iodosuccinimide (2.24 g, 9.94 x 10" 3 mol) was added in portions, the mixture warmed to 
room temperature, and stirred overnight. 19 F NMR analysis of the mixture showed no 
change in the spectrum from that of the starting material. The mixture was then heated 
at reflux for an additional 24 hours, at which time 19 F NMR analysis indicated no 
reaction. 

Preparation of 1.1.2. 2-Tetrafluoro-1 ,4-diiodobutane (105) 

Into a stainless steel pressure reactor of approximately 700 mL capacity was 
added 100 g (2.83 x 10" 1 mol) of 1,2-diiodotetrafluoroethane and 0.5 g (3.67 x 10 3 mol) 



114 

of d-limonene. The reactor was tightly sealed, connected to a vacuum line, immersed in 
a liquid nitrogen bath, and subjected to three successive freeze-pump-thaw cycles to 
remove oxygen. A total of 15.86 g (5.65 x 10~ 1 mol) of ethylene was transferred into the 
bomb, which was then sealed and placed into a heating manifold. (Caution: the 
following is performed behind a safety shield.) The thermostat was then set at 210° C, 
and with constant stirring the reaction mixture was heated for 8 hours. After cooling in 
an ice bath, venting, and opening, the contents (a dark violet liquid and dark solid) were 
diluted with 50 mL CHCI 3 , transferred to a 250 mL Erlenmeyer flask, and chilled in a 
freezer at -20° C, which caused precipitation of a large amount of dark solid. The 
mixture was filtered, the solid washed with three 20 mL portions of cold CHCI 3 , and the 
combined organic liquids rotary evaporated at room temperature. The remaining liquid 
was fractionally distilled at reduced pressure through a 15 cm Vigreux column. A total of 
43.36 g (40.1%) 105 was obtained as a violet liquid, bp 82-85° C / 22 mm Hg. 

1,1,2,2-Tetrafluoro-1,4-diiodobutane (105) : 1 H NMR: 8 2.70 (2H, m), 3.24 (2H, 
m); 13 C NMR: 8 -10.5, 34.7 (t, 2 J CF = 22.7 Hz), 96.7 (tt, 2 J CF = 42.8 Hz, 1 J CF = 315.9 Hz), 
116.5 (tt, 2 J CF = 30.9 Hz, 1 J CF = 253.1 Hz); 19 F NMR: f -60.8 (2F, t, 3 J FF = 4.9 Hz), -108.9 
(2F, tt, 3 J FF = 4.9 Hz, 3 J FH = 17.0 Hz). 

Preparation of 3.3.4.4-Tetrafluoro-4-iodobut-1-ene (106) 

Into a 250 mL three-necked round-bottomed flask equipped with magnetic stir 
bar, argon inlet, and pressure-equalizing addition funnel was added 30.69 g (8.03 x 10 
mol) 1,1,2,2-tetrafluoro-1,4-diiodobutane (105) dissolved in 60 mL anhydrous ether. 
DBU (26.88 g, 1.77 x 10~ 1 mol) dissolved in 50 mL anydrous ether was added dropwise 
at room temperature and allowed to stir for an additional 6 hours; at this time, 19 F NMR 
analysis of a small aliquot of the reaction mixture showed complete consumption of 
starting material. The mixture was poured into 50 mL of 5% aqueous HCI, the layers 
separated, and the aqueous phase extracted with three 10 mL portions of ether. The 



2 



115 

combined organic fractions were washed once with 20 ml_ of saturated NaHC0 3 
solution, once with 10 mL of water, dried over MgS0 4 , and the ether carefully removed 
by gentle, ambient pressure distillation through a 15 cm Vigreux column. Upon removal 
of most of the solvent in this way, the bath temperature was increased and 13.46 g 
(66.0%) 106 was collected over a range of 90-92° C. 

3.3.4.4-Tetrafluoro-4-iodobut-1-ene(106): 1 H NMR: 5 5.83 (1H, m), 5.99 (2H, m); 
13 C NMR: 5 97.3 (tt, 2 J CF = 44.1 Hz, 1 J CF = 316.5 Hz), 113.5 (tt, 2 J CF = 30.1 Hz, 1 J CF = 
249.9 Hz), 124.2 (t, 2 J CF = 26.1 Hz), 126.0 (t, 3 J CF = 8.6 Hz); 19 F NMR: * -60.7 (2F, t, 3 J FH 
= 7.3 Hz), -108.7 (2F, m). 

Preparation of 1,1,2,2-Tetrafluoro-1-iodobutane (107) 

Into a 250 mL three-necked round-bottomed flask equipped with reflux 
condenser, pressure-equalizing addition funnel and magnetic stir bar was added 10.0 g 
(3.94 x 10 2 mol) of 3,3,4,4-tetrafluoro-4-iodobut-1-ene (106) dissolved in 50 mL of dry 
methanol, 3.20 g (9.98 x 10" 2 mol) anhydrous hydrazine, and 0.1 g (1.01 x 10" 3 mol) 
cuprous chloride. The flask was cooled to 0° C, and 14.2 g of a 30% aqueous solution of 
hydrogen peroxide was delivered dropwise over a period of 20 minutes, over which time 
the color of the reaction mixture changed to powder blue then to a dark amber as the 
H 2 2 addition neared completion. The cooling bath was then removed, the mixture 
allowed to stir at room temperature for an additional 2 hours, and the contents poured 
into a solution of 1 mL of concentrated HCI in 200 mL of water. The mixture was 
extracted with six 10 mL portions of 1,2-dichlorobenzene, dried over Na 2 S0 4 , and 
transferred to a 100 mL flask attached to a 15 cm Vigreux column. A vacuum adapter 
attached to small trap (immersed in a dry ice-isopropanol slush) was attached, the 
pressure lowered to 100 mm Hg, and the flask heated with vigorous stirring until the 
contents began to boil. At this time, it was observed that approximately 1 mL of material 



116 

had accumulated in the trap, which was subjected to preparative GC separation. A total 
of 1.08 g (10.7%) 107 was collected as a colorless liquid. 

1.1.2.2-Tetrafluoro-1-iodobutane (107) : 1 H NMR: 8 1.15 (3H, t, 3 J HH = 7.5 Hz), 
2.11 (2H, m); 13 C NMR: 5.05, 23.0 (t, 2 J CF = 23.6 Hz); 98.5 (tt, 2 J CF = 43.5 Hz, 1 J CF = 
315.9 Hz); 117.5 (tt, 2 J CF = 29.6 Hz, 1 J CF = 250.9 Hz); 19 F NMR: <j) -59.9 (2F, t, 3 J FF = 4.9 
Hz), -1 1 0.4 (2F, tt, 3 J FF = 4.9 Hz, 3 J FH = 1 7. 1 Hz). 

Preparation of 1,1,2,2-Tetrafluorohexane (108) 

1-Bromo-1,1,2,2-tetrafluorohexane (102) (2.0 g, 8.44 x 10~ 3 mol) was dissolved in 
5 mL of benzene in a 25 mL round-bottomed flask equipped with septum-capped 
sidearm inlet and stir bar, and attached to a small distillation apparatus equipped with ice 
water condenser and fractionating column. The bath temperature was raised to 60° C 
and 2.95 g (1.01 x 10" 2 mol) nBu 3 SnH and 0.01 g (6.09 x 10" 5 mol) AIBN dissolved in 0.5 
mL benzene was added slowly via syringe through the rubber septum. When the 
addition was complete, the mixture was stirred for an additional 30 minutes and the bath 
temperature quickly raised to 150° C. All volatile material was flash distilled and 
subjected to preparative GC separation, which afforded pure 108 as a colorless liquid. 

1,1,2,2-Tetrafluorohexane (108): 1 H NMR: 5 0.95 (3H, t, 3 J HH = 7.5 Hz), 1.40 (2H, 
m), 1.55 (2H, m), 1.95 (2H, m), 5.70 (1H, tt, 3 J HF = 3.0 Hz, 2 J HF = 54.0 Hz); 13 C NMR: 
13.6, 22.4 (2C, overlapping), 29.6 (t, 2 J CF = 22.5 Hz), 110.4 (tt, 2 J CF = 41.1 Hz, 1 J CF = 
248.4 Hz), 118.1 (tt, 2 J CF = 28.5 Hz, 1 J CF = 243.8 Hz); 19 F NMR: <|> -116.8 (2F, t, 3 J FH = 
19.5 Hz), -136.1 (2F, d, 2 J FH = 53.7 Hz); HRMS for C 6 H 10 F 4 : calc. 158.0719, calc. (M-H) 
157.0640, found 157.0648. 

Preparation of 3.3.4.4-Tetrafluoro-1-phenvloctane (110) 

1-Bromo-1,1,2,2-tetrafluorohexane (102) (2.0 g, 8.44 x 10" 3 mol) and 1.76 g 
(1.69 x 10" 2 mol) styrene dissolved in 25 mL benzene were added to a 100 mL three- 
necked round-bottomed flask fitted with rubber septum, condenser with argon inlet, and 



117 

magnetic stirrer. A total of 4.90 g (1 .68 x 10" 2 mol) nBu 3 SnH and 0.05 g (3.04 x 10" 4 mol) 
AIBN were dissolved in 5 mL benzene and taken up in a syringe. A 150 W flood lamp 
was placed at a distance of approximately 15 cm from the flask, and with irradiation (at 
this distance, sufficient heat was generated to cause the solvent to reflux as well) the 
nBu 3 SnH solution was delivered to the reaction mixture via syringe pump over a 36 hour 
period. After rotary evaporation at elevated temperature, column chromatography (silica 
gel, hexanes) afforded four fractions of 110 free from organotin contaminants. 

3.3.4,4-Tetrafluoro-1-phenvloctane (110) : 1 H NMR: 5 0.96 (3H, t, 3 J HH = 7.2 Hz), 
1.41 (2H, m), 1.57 (2H, m), 2.02 (2H, m), 2.31 (2H, m), 2.90 (2H, m), 7.28 (5H, m) 13 C 
NMR: 1.0, 13.8, 22.5, 26.9, 29.7 (t, 2 J CF = 22.7 Hz), 32.1 (t, 2 J CF = 22.8 Hz), 118.8 (tt, 
2 Jcf = 37.6 Hz, 1 J CF = 250.4 Hz); 119.3 (tt, 2 J CF = 34.1 Hz, 1 J CF = 245.9 Hz), 126.3, 128.3, 
128.6, 140.3); 19 F NMR: <|> -116.0 (2F, m), -116.4 (2F, m); HRMS for C 14 H 18 F 4 : calc. 
262.1345, found 262.1307. 

Preparation of 1-rPerfluorohexvllethane (113) 

Into a 25 mL round-bottomed flask equipped with septum-capped sidearm inlet 
and magnetic stir bar was placed 4.22 g (8.90 x 10" 3 mol) 2-[perfluorohexyl]-1- 
iodoethane (112.) The flask was attached to a distillation apparatus equipped with a 
small fractionating column and ice water condenser. The bath temperature was raised 
to 80° C and 3.11 g (1.07 x 10~ 2 mol) nBu 3 SnH was slowly injected into the flask with 
stirring. After 30 minutes at this temperature, the bath temperature was raised to 150° C 
and the product distilled over a range of 81-82° C. Preparative GC purification afforded 
pure 113. 

HPerfluorohexvllethane (113). 1 H NMR: 5 1.14 (3H, m), 2.10 (2H, m); 19 F NMR: 
4> -81.5 (3F, t, 3 J FF = 9.8 Hz), -117.0 (2F, m), -122.5 (2F, br s), -123.4 (2F, br m), -124.2 
(2F, br m), -126.8 (2F, m); HRMS for C 8 H 5 F 13 : calc. 348.0183, calc. (M-F) 329.0200, 
found 329.0280. 






118 

Preparation of 2-lodo-1-fperfluorohexvH-4-phenylbutane (116) 

Into a 50 ml_ three-necked round-bottomed flask equipped with argon inlet, 
rubber septum and stir bar was added 0.5 g (3.78 x 10" 3 mol) 4-phenyl-1-butene and 
2.02 g (4.53 x 10~ 3 mol) perfluorohexyl iodide dissolved in 20 mL of hexanes. A 1.0 M 
solution of triethylborane in hexanes (0.4 mL, 4 x 10" 4 mol) was slowly injected through 
the septum, and the reaction allowed to stir for 6 hours at room temperature. The 
mixture was washed twice with 10 mL water, the solvent rotary evaporated, and the 
remaining liquid subject to reduced pressure fractional distillation. A total of 1.81 g 
(82.8%) 116 was collected as a light violet liquid, bp 125-127° C / 1 mm Hg. 

2-lodo-1-rperfluorohexvn-4-phenylbutane (116) : 1 H NMR: 5 2.12 (2H, m), 2.75 
(2H, m), 2.92 (2H, m), 4.27 (1H, m), 7.27 (5H, m); 19 F NMR: + -81.3 (3F, t, 3 J FF = 9.8 Hz), 
-111.8 (1F, dm, 2 J FF = 275.9 Hz), -115.2 (1F, dm, 2 J FF = 267.3 Hz), -122.3 (2F, br s), 
-123.3 (2F, s), -124.2 (2F, br s), -126.7 (2F, m); HRMS for C 16 H 12 F 13 I: calc. 577.9776, 
found 577.9815. 

Preparation 1-fPerfluorohexvl1-4-phenylbutane (117) 

Into a 50 mL three-necked round-bottomed flask equipped with argon inlet and 
magnetic stirrer was added 1.5 g (2.59 x 10" 3 mol) 2-iodo-1-[perfluorohexyl]-4- 
phenylbutane (116) dissolved in 10 mL DMSO. The bath temperature was raised to 
70° C and 0.39 g (1.03 x 10~ 2 mol) sodium borohydride was added in portions to the 
reaction mixure. Stirring was continued for an additional 6 hours, at which time the 
mixture was poured into 50 mL of ice water, and carefully acidified with 6 M HCI. 10 mL 
ether was added, the layers separated, and the aqueous layer extracted with three 10 
mL portions of ether. The combined organic layers were washed once with 10 mL 5% 
aqueous NaHC0 3 and twice with 10 mL of water, dried over MgS0 4 , and the solvent 
rotary evaporated. Reduced pressure fractional distillation afforded 0.82 g (70.0 %) 117, 
bp 83-85° C / 0.25 mm Hg. 






119 

1-rPerfluorohexvll-4-phenvlbutane (117) : 1 H NMR: 8 1.67 (4H, m), 2.06 (2H, m), 
2.64 (2H, t, 3 J HH = 6.9 Hz), 7.24 (5H, m); 19 F NMR: <)> -81.4 (3F, t, 3 J FF = 9.8 Hz), -114.9 
(2F, m), -122.5 (2F, br s), -123.4 (2F, s), -124.1 (2F, s), -126.7 (2F, m); HRMS for 
Ci 6 H 13 F 13 : calc. 452.0810, found 452.0797. 

Preparation of Pentafluoroethane (120) 

lodopentafluoroethane (2.16 g, 8.78 x 10" 3 mol) was condensed into a Pyrex tube 
equipped with a Rotaflo stopcock, rubber septum, and immersed in a dry ice-isopropanol 
slush. In one portion, 3.07 g (1.05 x 10" 2 mol) nBu 3 SnH was injected into the tube and 
the stopcock closed. The tube was then subjected to photolysis in a Rayonet reactor for 
30 minutes at room temperature with periodic shaking. The tube was cooled to -20° C in 
a dry ice-isopropanol bath, a rubber hose connected to the Rotaflo tube and to a trap 
immersed in a liquid nitrogen-ether slush, and the stopcock opened. After trap-to-trap 
transfer in this way, an NMR tube containing ca. 1 mL CDCI 3 , capped with a rubber 
septum, and immersed in a liquid-nitrogen-ether slush was connected via cannula. The 
trap was warmed to -20° C as a sample of 120 collected in the tube, which was flame 
sealed and taken for NMR analysis. 

Pentafluoroethane (120) : 1 H NMR: 5 5.88 (tq, 3 J HF = 2.4 Hz, 2 J HF = 52.2 Hz); 19 F 
NMR: <)> -86.1 (3F, s), -138.5 (2F, d, 2 J FH = 51 .3 Hz). 

Preparation of 1.1.1. 2.2-Pentafluoro-4-phenvlbutane (122) 

lodopentafluoroethane (5.47 g, 2.22 x 10" 2 mol) of was condensed into a Pyrex 
Rotaflo tube containing 2.12 g (2.04 x 10" 2 mol) styrene, 5.92 g (2.03 x 10" 2 mol) 
nBu 3 SnH, 5 mL benzene and a small sir bar. The stopcock was closed and the tube 
irradiated with a 150 W flood lamp with stirring for 72 hours. Volatiles were removed by 
rotary evaporation and the residue subject to flash vaccuum distillation at 100 mm Hg 
(bath temperature 150° C) during which time the distillate temperature reached 69° C. 
Preparative GC purification of the distillate yielded an analytically pure sample of 122. 



120 

1.1.1 .2.2-Pentafluoro-4-phenvlbutane (1 22): 1 H NMR: 5 2.38 (2H, m), 2.95 (2H, 
m), 7.31 (5H, m); 13 C NMR: 5 26.6, 32.7 (t, 2 J CF = 21.5 Hz), 115.5 (tq, 2 J CF = 37.6 Hz, 1 J CF 
= 250.4 Hz), 119.3 (qt, 2 J CF = 36.1 Hz, 1 J CF = 283.4 Hz), 126.7, 128.2, 128.8, 139.2; 19 F 
NMR: + -85.9 (3F, s), -119.1 (2F, t, 3 J FH = 19.5 Hz); HRMS for C 10 H 9 F 5 : calc. 224.0624, 
found 224.0670; CHN for C 10 H 9 F 5 : calc. 53.58% C, 4.05% H, found 53.42%C, 3.97% H. 

Preparation of 1-(tert-Butvldimethylsiloxvl)-3-butene (130) 

Into a 250-mL three-necked round-bottomed flask equipped with a condenser 
and argon inlet was placed 9.60 g (1.33 x 10 " 1 mol) of 3-buten-1-ol, 20 mL DMF, 24.1 g 
(1.60 x 10" 1 mol) terf-butyldimethylsilyl chloride, and 22.7 g (3.33 x 10" 1 mol) imidazole. 
This was stirred for 48 hours at room temperature under an argon atmosphere. The 
contents of the flask were then poured into 250 mL of pentane, and washed with three 
50 mL portions of water followed by three 50 mL portions of saturated aqueous sodium 
chloride. The organic phase was dried, the solvent rotary evaporated, and the resulting 
liquid subjected to fractional reduced pressure distillation through a 15 cm vigreaux 
column. A total of 22.08 g (89.3%) of pure 130 was obtained in 4 fractions as a colorless 
liquid boiling at 102-105° C / 75 mm Hg. 

1 -(tert-Butvldimethvlsiloxyl)-3-butene (1 30): 1 H NMR (CDCI 3 ): 8 0.05 (6H, s), 0.90 
(9H, s), 2.27 (2H, dt, 3 J HH = 7 Hz, 3 J HH = 3 Hz), 3.66 (2H, t, 3 J HH = 7 Hz), 5.02 (1H, m), 
5.10 (1H, m), 5.81 (1H, m); 13 C NMR (CDCI 3 ): 5 -5.27, 18.3, 25.9, 37.5, 62.8, 116.2, 
135.4; HRMS for C 10 H 22 SiO: calc. 186.1440, found (M+H) 187.1561; CHN for C 10 H 22 SiO: 
calc. 64.45% C, 11.90% H, found 64.32%C, 12.03 % H. 

Preparation of 1.3-Dibromo-5-(tert-butvldimethylsiloxvl)-1,1-difluoropentane (131) 

Cuprous chloride (0.10 g, 1.25 x 10~ 3 mol), along with 12.5 mL terf-butanol, 
3.83 g (6.26 x 10" 2 mol) ethanolamine, 23.30 g (1.25 x 10 1 mol) 1-(terf- 
butyldimethylsiloxyl)-3-butene (130), and 52.60 g (2.51 x 10 _1 mol) CF 2 Br 2 were added 



121 

to a Carius tube. A small stir bar was added and the tube flame-sealed. After stirring at 
85° C for 96 hours (performed behind a safety shield) the tube was cooled in an ice bath, 
opened, and the contents transferred to a 500 ml_ Erlenmeyer flask. The tube was 
rinsed with four 50 ml_ portions of hexanes, and the combined organic material filtered 
through a 50 mL pad of silica gel, which was rinsed with three additional 50 mL portions 
of hexanes. Rotary evaporation of the solvent afforded a colorless liquid judged by 1 H 
NMR to contain some unreacted starting material, 2.78 g of which was successfully 
recovered by reduced pressure distillation at 102-105° C / 75 mm Hg. High vacuum was 
then applied and a total of 35.83 g (72.4%, 82.0% based on consumed 130) 131 was 
obtained as a colorless liquid boiling at 75-79° C / 0.09 mm Hg. 

1 ,3-Dibromo-5-(fert-butvldimethvlsiloxyl)-1 , 1 -difluoropentane (1 31 ): 1 H NMR: 5 
0.07 (3H, s), 0.08 (3H, s), 0.90 (9H, s), 1.95 (1H, m), 2.15 (1H, m), 3.08 (2H, m), 3.80 
(2H, m), 4.46 (1H, m); 13 C NMR: 8 -5.5, 18.2, 25.9, 41.3, 43.8, 52.9 (t, 2 J CF = 19 Hz), 
60.2, 120.64 (t, 1 J CF = 306 Hz); 19 F NMR: f -43.1 (m); HRMS for CnH^SiOFzBr;,: calc. 
393.9774, calc. (M-f-C 4 H 9 ) 336.9070, found 336.905; CHN for C 11 H 22 SiOF 2 Br 2 : calc. 
33.35% C, 5.60% H, found 33.62% C, 5.62% H. 

Preparation of 1-Bromo-5-(terf-butyldimethvlsiloxvl)-1.1 -difluoropentane (132) 

1,3-Dibromo-5-(ferf-butyldimethylsiloxyl)-1,1 -difluoropentane (131) (30.15 g, 7.60 
x 10' 2 mol) was dissolved in 150 mL of dry DMSO in a 500 mL three-necked round- 
bottomed flask equipped with an argon inlet and strong magnetic stir bar. Sodium 
borohydride (11.5 g, 3.04 x 10" 1 mol) was then added in portions with vigorous stirring. 
After the addition was complete, the temperature was raised to 70-75° C over the course 
of one hour and stirring continued for an additional 6 hours, at which time analysis of the 
reaction mixture by 19 F NMR demonstrated complete consumption of starting material. 
The flask was cooled and carefully quenched with ca. 100 g of ice, and the contents 
carefully acidified with concentrated hydrochloric acid and transferred to a 1 liter 



122 

separatory funnel. After extraction with three 100 mL portions of ether, the combined 
extracts were washed with two 25 mL portions of water, dried over MgS0 4 , and rotary 
evaporated. The remaining liquid was distilled at reduced pressure through a 15 cm 
vigreaux column, affording 21.70 g (90.0%) 132 as a colorless liquid, bp 108-111° C / 10 
mm Hg. 

1-Bromo-5-(fert-butvldimethvlsiloxvl)-1,1-difluoropentane (132): 1 H NMR: 8 0.05 
(6H, s), 0.90 (9H, s), 1.63 (4H, m), 2.38 (2H, m), 3.63 (2H, t, 3 J HH = 6 Hz); 13 C NMR: 8 
18.3, 20.7, 25.9 (2C, overlapping), 31.4, 44.1 (t, 2 J CF = 22.5 Hz), 62.4, 123.2 (t, 1 J CF = 
304 Hz); 19 F NMR: + -44.0 (m); HRMS for d^SiC^Br: calc. 316.0669, found (M+H) 
317.0630. 

Preparation of 5-Bromo-5,5-difluoropentan-1-ol (133) 

Into a 250-mL round-bottomed flask was placed 15.04 g (4.74 x 10~ 2 mol) 
1-bromo-5-(ferf-butyldimethylsiloxyl)-1,1-difluoropentane (132) along with 41 mL of 
acetonitrile. To this was slowly added with stirring 7.70 g (4.75 x 10~ 2 mol) ferric chloride. 
The reaction mixture turned a brick-red color and became slightly warm. The reaction 
was allowed to stir for 3 hours at room temperature, at which time the contents of the 
flask were poured into 250 mL of water and 100 mL of chloroform was added. The 
chloroform layer was drained and the aqueous layer extracted with three 50 mL portions 
of chloroform. These combined extracts were washed twice with 25 mL of water, dried 
over MgS0 4 and the solvent rotary evaporated. A total of 9.41 g (97.7%) 132 was 
collected by fractional reduced pressure distillation, bp 80-82° C / 10 mm Hg. 

5-Bromo-5.5-difluoropentan-1-ol (133) : 1 H NMR: 8 1.39 (1H, s), 1.69 (4H, m), 
2.40 (2H, m), 3.69 (2H, t, 3 J HH = 6 Hz); 13 C NMR: 8 20.5, 31.3, 44.0 (t, 2 J CF = 21.5 Hz), 
62.2, 123.0 (t, 1 J CF = 303.5 Hz); 19 F NMR: * -44.1 (t, 3 J FH = 14.7 Hz); HRMS for 
C 5 H 9 F 2 BrO: calc. 201.9804, found (M+H) 202.9957; CHN for C 5 H 9 F 2 BrO: calc. 29.58% 
C, 4.47% H, found 29.71% C, 4.46% H. 



123 

Preparation of 5-Bromo-5,5-difluoropentanal (134) 

Into a 250-mL round-bottomed flask equipped with a magnetic stir bar was added 
8.5 g (4.18 x 10~ 2 mol) 5-bromo-5,5-difluoropentan-1-ol (133) dissolved in 85 mL of 
dichloromethane. PCC (13.53 g, 6.28 x 10" 2 mol) was added slowly in portions with 
vigorous stirring. After the addition was complete, the mixture was allowed to stir at 
room temperature for an additional 6 hours. The darkened reaction mixture (which 
demonstrated complete consumption of starting material by TLC analysis) was filtered 
through a pad of silica gel, which was rinsed with an additional three 10 mL portions of 
CH 2 CI 2 . Rotary evaporation of the solvent followed by fractional distillation at reduced 
pressure afforded 4.46 g (53.1%) 134 as a colorless liquid, bp 100-102° C / 50 mm Hg. 

5-Bromo-5.5-difluoropentanal (134): 1 H NMR: 5 1.97 (2H, m), 2.41 (2H, m), 2.59 
(2H, t, 3 J HH = 7.2 Hz), 9.80 (1H, s); 13 C NMR: 5 16.6 (t, 3 J CF = 3.5 Hz), 42.1, 43.2 (t, 2 J CF = 
22.0 Hz); 122.5 (t, 1 J CF = 303.4 Hz), 200.5; 19 F NMR: * -44.4 (t, 3 J FH = 14.4 Hz); HRMS 
for C 5 H 7 F 2 BrO: calc. 199.9648, found (M+H) 200.9726. 

Preparation of 6-Bromo-6.6-difluorohex-1-ene (135) 

A 100 mL three-necked round-bottomed flask equipped with self-equalizing 
addition funnel, argon inlet and magnetic stir bar was charged with 9.12 g (2.55 x 10 
mol) of methyltriphenylphosphonium bromide and 20 mL anhydrous THF. The flask was 
cooled to 0° C and 9.4 mL of a 2.5 M solution of butyllithium in hexanes (2.35 x 10" 2 mol) 
was added dropwise. After addition was complete, the mixture was stirred for an 
additional 30 minutes at 0° C. 5-Bromo-5,5-difluoropentanal (134) (4.28 g, 2.13 x 10 
mol) was dissolved in 20 mL anhydrous THF and added dropwise to the reaction 
mixture. After addition the mixture was allowed to warm to room temperature and stir for 
an additional 6 hours. The contents were poured into 50 mL of water and extracted with 
five 20-mL portions of ether. The combined ether fractions were dried over MgS0 4 , 
filtered, and the solution concentrated by distillation through a 15 cm vigreaux column. 



2 



-2 



124 

Upon removal of most of the ether and residual THF the product was distilled at ambient 
pressure, yielding 2.04 g (48.1%) 135, bp 120-123° C. An analytically pure sample was 
obtained by preparative GC for spectroscopic analysis and kinetic experiments. 

6-Bromo-6.6-difluorohex-1-ene (135) : 1 H NMR: 8 1.74 (2H, m), 2.15 (2H, 
overlapping dt, J = 7 Hz), 2.35 (2H, m), 5.05 (2H, m), 5.77 (1H, m); 13 C NMR: 5 23.1, 
32.3, 43.6 (t, 2 J CF = 21.6 Hz), 115.9, 123.1 (t, 1 J CF = 303.5 Hz), 137.0; 19 F NMR: <|> -43.9 
(t, 3 J FH = 14.7 Hz); HRMS for C 6 H 9 F 2 Br: calc. 197.9855, calc (M-Br) 119.0672, found 
1 19.0667; CHN for C 5 H 9 F 2 Br: calc. 36.21% C, 4.56% H, found 36.28% C, 4.56% H. 

Preparation of 6,6-Difluorohex-1-ene (136) 

6-Bromo-6,6-difluorohex-1-ene (135) (1.0 g, 5.02 x 10" 3 mol) was dissolved in 
mesitylene (1 mL) in a 10 mL round-bottomed flask equipped with a septum-capped 
sidearm inlet and small stir bar. This was attached to an ice-water-cooled micro 
distillation apparatus. Tributyltin hydride (1.6 g, 5.50 x 10" 3 mol) was slowly injected into 
the flask through the septum. When the addition was complete, heating was begun with 
an oil bath. After 15 minutes at 50° C, the temperature was quickly raised and all volatile 
material was flash distilled until the bath temperature reached 150° C. The distillate was 
subjected to preparative GC separation, affording pure 136. 

6,6-Difluorohex-1-ene (136) : 1 H NMR: 8 1.57 (2H, m), 1.83 (2H, m), 2.12 (2H, 
overlapping dt, J = 7 Hz), 5.00 (2H, m), 5.79 (1H, m), 5.81 (1H, tt, 3 J HH = 4 Hz, 2 J HF = 57 
Hz); 13 C NMR: 8 21.3, 32.9, 33.8 (t, 2 J CF = 20.5 Hz), 115.4, 117.3 (t, 1 J CF = 237.6 Hz), 
120.5; 19 F NMR: + -116.4 (dt, 3 J FH = 14.6 Hz, 2 J FH = 59.8 Hz); HRMS for C 6 H 10 F 2 : calc. 
120.0751, found 120.0756; CHN for C 6 H 10 F 2 : calc. 59.98% C, 8.39% H, found 59.96% C, 
8.47% H. 

Preparation of 1.1-Difluoro-2-methylcvclopentane (138) 

Into a 50 mL three-necked round-bottomed flask equipped with magnetic stirrer 
and septum was placed 0.5 g (5.09 x 10~ 3 mol) 2-methylcyclopentanone dissolved in 10 



125 

mL of anhydrous CH 2 CI 2 . DAST (0.9 g, 5.58 x 10~ 3 mol) was then injected and the 
mixture stirred at room temperature overnight. The reaction was dispensed onto ca. 2 g 
of ice, the layers separated, and the organic layer washed with 1 mL of saturated 
aqueous NaHC0 3 . After drying, all volatile material was flash distilled and subjected to 
preparative GC, affording pure 138. 

1 . 1 -Difluoro-2-methvlcvclopentane (1 38): 1 H NMR: 5 1.04 (3H, d, 3 J HH = 7 Hz), 
1.40 (1H, m), 1.73 (2H, m), 2.04 (4H, overlapping m); 13 C NMR: 5 12.1, 19.9, 30.9, 34.4 
(t, 2 J CF = 25.1 Hz), 40.7 (t, 2 J CF = 23.5 Hz), 132.4 (t, 1 J CF = 249.4 Hz); 19 F NMR: <j> -100.2 
(1F, d of overlapping dt, 3 J FH = 12.2 Hz, 2 J FF = 225.8 Hz), -107.6 (1F, d of overlapping dt, 
3 J FH = 17.1 Hz, 2 J FF = 224.6 Hz); HRMS forC 6 H 10 F 2 : calc. 120.0751, found 120.0748. 

Preparation of 6-Bromohex-1-en-5-ol (142) 

To a 500 mL three-necked round-bottomed flask equipped with magnetic stirrer 
was added 100 mL acetic acid, 50 mL of saturated aqueous potassium bromide, and 50 
mL THF. The flask was cooled to 0° C and 5.0 g (5.09 x 10" 2 mol) of 1,2-epoxy-5- 
hexene dissolved in 10 mL of THF was added dropwise with stirring. The 
heterogeneous mixture was stirred at 0° C for an additional two hours, then allowed to 
warm to room temperature and stir overnight. Most of the THF was removed by rotary 
evaporation, 100 mL of ether and 50 mL of water was added, and the aqueous layer 
washed with saturated aqueous NaHC0 3 until the acetic acid was removed. Drying over 
MgS0 4 followed by rotary evaporation of the solvent afforded 8.03 g (88.1%) 142 which 
was used in the next step without further purification. 

6-Bromohex-1-en-5-ol (142). 1 H NMR: 5 1.61 (2H, overlapping dt, J = 8 Hz), 2.15 
(2H, m), 2.67 (1H, s), 3.35 (1H, m), 3.48 (1H, m), 3.76 (1H, m), 4.99 (2H, m), 5.77 (1H, 
m); 13 C NMR: 5 29.6, 34.0, 40.0, 70.2, 115.2, 137.5; HRMS for CeHnBrO: calc. 
177.9993, calc (M+H) 178.9993, found 179.0058. 



126 

Preparation of 1-Bromo-5-hexen-2-one (143) 

6-Bromohex-1-en-5-ol (142) (7.25 g, 4.05 x 10" 2 mol) dissolved in 10 mL ether 
was added dropwise to a mixture of 60 mL of Jones' reagent and 25 mL ether at room 
temperature with stirring. After 4 hours the dark green reaction mixture was diluted with 
50 mL of water. The layers were separated and the aqueous layer extracted with three 
20 mL portions of ether. The combined organic extracts were washed twice with 20 mL 
of saturated aqueous NaHC0 3 and once with 20 mL of water. Drying and rotary 
evaporation of the solvent afforded 5.92 g (82.6%) 143 which was used without further 
purification. 

1-Bromo-5-hexen-2-one (143) : 1 H NMR: 5 2.35 (2H, overlapping dt, J = 6 Hz), 
2.74 (2H, t, 3 J HH = 7 Hz), 3.88 (2H, s), 5.01 (2H, m), 5.78 (1H, m); 13 C NMR: 5 27.7, 34.2, 
38.8, 115.7, 136.3, 201.2; HRMS for C 6 H 9 BrO: calc. 175.9836, found 175.9850. 

Preparation of 6-Bromo-5,5-difluorohex-1-ene (144) 

A 100 mL three-necked round-bottomed flask equipped with an argon inlet, 
rubber septum, and magnetic stirrer was charged with 2.1 g (1.19 x 10" 2 mol) 1-bromo-5- 
hexen-2-one (143) in 20 mL of anhydrous CH 2 CI 2 . The flask was cooled to 0° C and 1 .9 
mL (2.32 g, 1.44 x 10~ 2 mol) of diethylaminosulfurtrifluoride was slowly injected into the 
reaction mixture with stirring. After two hours at 0° C, the flask was allowed to warm to 
room temperature and stirring continued for an additional 48 hours. The contents were 
carefully dispensed onto 20 g of ice, the layers separated, and the aqueous layer 
extracted twice with 5 mL CH 2 CI 2 . The combined organic extracts were washed once 
with 10 mL of saturated aqueous NaHC0 3 and once with 10 mL of water. After drying 
over MgS04 the solution was carefully concentrated via gentle ambient pressure 
distillation. A total of 1.33 g (56.2%) 144 was obtained as a colorless liquid, bp 
117-119° C, which was further purified by preparative GC for spectroscopic analysis and 
kinetic experiments. 



127 

6-Bromo-5.5-difluorohex-1-ene(144): 1 H NMR: 5 2.06 - 2.31 (4H, m), 3.53 (2H, t, 
3 J HF = 13 Hz), 5.07 (2H, m), 5.82 (1H, m); 13 C NMR: 5 26.2 (t, 3 J CF = 4.5 Hz), 31.3 (t, 2 J CF 
= 33.6 Hz), 33.8 (t, 2 J CF = 24.1 Hz), 115.8, 121.1 (t, 1 J CF = 241.4 Hz), 136.2; 19 F NMR: <|> 
-99.3 (m); HRMS for C 6 H 9 F 2 Br: calc. 197.9855, found 197.9850; CHN for C 6 H 9 F 2 Br: calc. 
36.21% C, 4.56% H, found 36.16% C, 4.57% H. 

Preparation of 5,5-Difluorohex-1-ene (144) 

6-Bromo-5,5-difluoro-1-hexene (144) (1.0 g, 5.02 x 10 3 mol) was treated with 
1.6 g (5.50 x 10" 3 mol) of tributyltin hydride in a manner identical to the independent 
preparation of 136. Flash distillation followed by preparative GC separation afforded 
pure 144. 

5,5-Difluorohex-1-ene (144): 1 H NMR: 5 1.60 (3H, t, 3 J HF = 18 Hz), 1.94 (2H, m), 
2.24 (2H, m), 5.03 (2H, m), 5.83 (2H, m); 13 C NMR: 8 23.3 (t, 2 J CF = 28.1 Hz), 26.9 (t, 
3 J CF = 5.0 Hz), 37.2 (t, 2 J CF = 25.1 Hz), 115.2, 123.9 (t, 1 J CF = 236.4 Hz), 136.9; 19 F NMR: 
(j) -91.3 (m); HRMS for C 6 H 10 F 2 : calc. 120.0751, found 120.0743. 

Preparation of 1,1-Difluoro-3-methylcvclopentane (147) 

3-Methylcyclopentanone (0.5 g, 5.09 x 10 3 mol) and 0.9 g (5.58 x 10" 3 mol) 
DAST were reacted in 10 mL of anhydrous CH 2 CI 2 in a manner identical to the 
preparation of 138. Flash distillation and preparative GC separation afforded pure 147. 

1 ,1-Difluoro-3-methvlcvclopentane (147): 1 H NMR: 5 1.05 (3H, d, 3 J HH = 6Hz), 
1 .36 (1 H, m), 1 .61 (1 H, m), 1 .87 - 2.31 (5H, m); 13 C NMR: 5 20.0, 31 .6, 32.0 (t, 3 J CF = 4.3 
Hz), 36.0 (t, 2 J CF = 25.0 Hz), 44.0 (t, 2 J CF = 23.6 Hz), 133.0 (t, 1 J CF = 246.9 Hz); 19 F NMR: 
+ -88.9 (1F, dm, 2 J FF = 217.1 Hz), -90.2 (1F, dm, 2 J FF = 227.1 Hz); HRMS for C 6 H 10 F 2 : 
calc. 120.0751, found 120.0759. 



128 

Preparation of 1,1-Difluorocyclohexane (149) 

Cyclohexanone (0.5 g, 5.09 x 10" 3 mol) and 0.9 g (5.58 x 10~ 3 mol) DAST were 
reacted in 10 mL of anhydrous CH 2 CI 2 in a manner identical to the preparation of 138. 
Preparation and flash distillation afforded pure 149. 

1.1-Difluorocvclohexane (149) : 1 H NMR: 5 0.97 (2H, m), 1.29 (4H, q, 3 J HH = 6 
Hz), 1.58 (4H, m); 13 C NMR: 5 22.8, 24.4, 34.1 (t, 2 J CF = 23.5 Hz), 123.6 (t, 1 J CF = 239.9 
Hz); 19 F NMR: <)> -95.7 (2F, br s). 

Preparation of 5,5,6, 6-Tetrafluorohex-1-ene (153), 1,1,2,2-Tetrafluoro-3-methylcvclo- 
pentane (155), and 1,1,2,2-Tetrafluorocyclohexane (157) 

6-Bromo-5,5,6,6-tetrafluorohex-1-ene (101) (5.0 g, 2.13 x 10~ 2 mol) dissolved in 5 
mL of mesitylene was added to a 50 mL three-necked round-bottomed flask equipped 
with ice water condenser, argon inlet, magnetic stir bar and rubber septum. Tributyltin 
hydride (7.5 g, 2.58 x 10" 2 mol) and 0.05 g (3.04 x 10" 4 mol) 2,2'-azobisisobutyronitrile 
(AIBN) in 5 mL of mesitylene were taken up into a syringe. The flask was heated at 
50° C and irradiated with a 150 W flood lamp placed at a distance of ca. 1 m while the 
nBu 3 SnH solution was delivered to the reaction mixture, via syringe pump, over a 24 
hour period. After the addition was complete, all volatile material was flash distilled from 
the reaction mixture until the bath temperature reached 150° C. Purification by 
preparative GC afforded pure samples of 153, 155, and 157. 

5,5,6.6-Tetrafluorohex-1-ene (153): 1 H NMR: 5 2.06 (2H, m), 2.33 (2H, m), 5.08 
(2H, m), 5.72 (1H, tt, 3 J HF = 3 Hz, 2 J HF = 54 Hz), 5.84 (1H, m); 13 C NMR: 5 24.6 (t, 3 J CF = 
4.0 Hz), 29.2 (t, 2 J CF = 22.1 Hz), 110.3 (tt, 2 J CF = 41.1 Hz, 1 J CF = 247.7 Hz), 115.9, 117.8 
(tt, 2 J CF = 29.0 Hz, 1 J CF = 244.8 Hz), 136.1; 19 F NMR: <f> -116.7 (2F, t, 3 J FH = 17.1 Hz), 
-136.0 (2F, d, 2 J FH = 56.2 Hz); HRMS for C 6 H 8 F 4 : calc. 156.0562, found 156.0562. 

1,1,2,2-Tetrafluoro-3-methvlcvclopentane (155): 1 H NMR 5 1.12 (3H, d, 3 J HH = 7 
Hz), 1.47 (1H, m), 2.00 (1H, m), 2.07 - 2.49 (3H, m); 13 C NMR: 8 11.4, 23.7 (m), 29.8 (t, 
2 J CF = 22.8 Hz), 36.2 (t, 2 J CF = 21.0 Hz), 117.6 - 125.8 (2C, m); 19 F NMR: + -110.1 (1F, 



129 

dm, 2 J FF = 234.4 Hz), -120.7 (1F, dm, 2 J FF = 239.3 Hz), -126.0 (1F, dt, 3 J FH = 12.2 Hz, 
2 J FF = 236.8 Hz), -132.9 (1F, dm, 2 J FF = 235.6); HRMS for C 6 H 8 F 4 : calc. 156.0562, found 
156.0563; CHN for C 6 H 8 F 4 : calc. 46.16% C, 5.16% H, found 46.17% C, 5.35% H. 

1 . 1 .2.2-Tetrafluorocyclohexane (1 57): 1 H NMR: 5 1.69 (4H, br s), 2.06 (4H, br s); 
13 C NMR: 5 21.0, 31.7 (t, 2 J CF = 22.1 Hz), 117.0 (tt, 2 J CF = 28.1 Hz, 1 J CF = 250.4 Hz); 19 F 
NMR: t -1 1 9.7 (4F, br s); HRMS for C 6 H 8 F 4 : calc. 1 56.0562, found 1 56.0571 . 

Competitive Kinetic Procedures 

Competition Kinetics: Hydrogen Atom Abstraction (k n ) versus Addition (k^ and 
Hydrogen Atom Abstraction versus Cyclization (k en )- General Procedure. 

Into each of a set of six Pyrex NMR tubes were added a known amount of C 6 D 6 , 

varying, known amounts of trapping agent or trapping agent and styrene, and a known 

amount of trifluorotoluene as an internal 19 F NMR standard. Each tube was sealed with 

rubber septa secured with PTFE tape, frozen in a dry ice-isopropanol slush, and 

subjected to three successive freeze-pump-thaw cycles followed by pressurization with 

argon. Into each frozen tube was then injected a known amount of the radical precursor 

(in the case of 119, a stock solution in degassed C 6 D 6 ) followed by warming to room 

temperature (in the case of 119, the tubes were flame-sealed before warming) with 

vigorous shaking. The tubes were then subjected to UV photolysis in a Rayonet reactor 

(254 nm lamps) until complete consumption of starting material was demonstrated by 19 F 

NMR analysis. Product ratios for varied concentrations of trapping agent or ratios of 

trapping agent to styrene allow determination of the ratios k H I k Cn or k H I k aM . Yields are 

determined by integration of product resonances versus that of internal standard (<j> 

-63.24) in the 19 F NMR. 

1,1-Difluorohex-1-vl Radical (77^ 

Ratios of [81] / [83] were determined by integration of the -CF 2 H and -CF 2 - 
resonances (at f -116.0 and -99.1, respectively) in the 19 F NMR. 



130 



Table 5-1. Data for Competitive Determination of k H (nBu 3 SnH) / /c add (Styrene) for 1,1 - 
Difluorohex-1-yl Radical (77). 



0.094 
0.094 
0.094 
0.094 
0.094 
0.094 



,CH=< 


3H? 1 


r nBu.SnH 1 / FC R H S CH=CH, 1 
0.719 


r 81 1 / [ 83 1 
2.30 


% Yield 


2.01 




95 


1.81 




0.847 


2.73 


95 


1.61 




1.01 


3.26 


96 


1.41 




1.21 


3.93 


97 


1.21 




1.49 


4.88 


96 


1.01 


k H 


1.87 
Slope = k H 1 /c add = 3.39 ± 0.02 
(nBu 3 SnH) = 9.1 (+ 1.7) x 10 6 M" 1 s" 1 


6.21 

i 


95 



2,2-Difluorohex-1-yl Radical (88) 

Ratios of [93] / [98] were obtained by integration of the respective -CF 2 - 
resonances at <j> -91 .0 and -98.6. 

Table 5-2. Data for Competitive Determination of k H (nBu 3 SnH) / k aM (Styrene) for 2,2- 
Difluorohex-1-yl Radical (88). 



[92] 


[ C„H fi CH=CH, 1 


r nBu.SnH 1 / rC fi H,CH=CH, 1 


[ 931 / f 98 1 


% Yield 


0.087 


4.08 


0.223 


5.96 


96 


0.087 


3.97 


0.243 


6.62 


98 


0.087 


3.85 


0.263 


7.16 


97 


0.087 


3.73 


0.285 


7.79 


98 


0.087 


3.62 


0.308 


8.33 


97 


0.087 


3.50 


0.333 


9.01 


100 



Slope = k H I k aM = 27.3 ± 0.6 
k H (nBu 3 SnH) = 1.4 (± 0.5) x 10 7 M' 1 s" 1 



131 

1.1,2,2-Tetrafluorohex-1-yl Radical (100) 

Ratios of [108] / [110] were obtained by the sum of integrals of the -CF 2 - and 
-CF 2 H resonances of 108 (at <|> -117.1 and -136.1, respectively) versus that of the two 
-CF 2 - resonances of 110 (at f -116.0 and -116.4, respectively). 



Table 5-3. Data for Competitive Determination of k H (nBu 3 SnH) / k aM (Styrene) for 1,1- 
2,2-Tetrafluorohex-1-yl Radical (100). 

r 102 1 f C fi H fi CH=CH, 1 f nBu . SnH 1 / rC fi H s CH=CH ? 1 f 1081/ [1101 % Yield 



0.085 


1.86 






0.506 


2.33 


98 


0.085 


1.68 






0.611 


2.90 


97 


0.085 


1.49 






0.741 


3.56 


97 


0.085 


1.30 






0.909 


4.30 


97 


0.085 


1.12 






1.13 


5.34 


97 


0.085 


0.931 


Ah 


Slope = /c H 
(nBu 3 SnH) = 


1.45 
1 k aM = 4.56 ±0.10 
9.2(±0.8)x10 7 M- 1 s" 1 


6.65 


99 



Table 5-4. Data for Competitive Determination of k H ((TMS) 3 SiH) / k aM (Styrene) for 1,1- 
2,2-Tetrafluorohex-1-yl Radical (100). 

r 102 



0.074 


1.29 








0.988 




1.10 


90 


0.074 


1.19 








1.11 




1.20 


92 


0.074 


1.08 








1.26 




1.32 


93 


0.074 


0.970 








1.44 




1.52 


94 


0.074 


0.863 








1.67 




1.71 


92 


0.074 


0.755 








1.96 




1.98 


95 








Slope = 


■■k H /k 


add = 0.913 + 0.017. 










*H 


((TMS) 3 SiH) = 


1.8(±0.1)x10 7 


M- 1 s- 1 







132 

2-fPerfluorohexvneth-1-yl Radical (111) 

Ratios of [113] / [117] were determined by integration of the respective -CF 2 - 
resonances at + -1 1 7. 1 and -115.0. 



Table 5-5. Data for Competitive Determination of k H (nBu 3 SnH) / k aM (Styrene) for 2- 
[Perfluorohexyl]eth-1-yl Radical (111). 

f 1121 [ C fi H s CH=CH, 1 f nBu . SnH 1 / fC gH g CH=CH ? 1 [1131/f 1171 % Yield 



0.064 


2.96 




0.217 




3.11 


100 


0.064 


2.83 




0.248 




3.58 


100 


0.064 


2.69 




0.283 




4.13 


100 


0.064 


2.56 




0.320 




4.72 


100 


0.064 


2.42 




0.362 




5.44 


100 


0.064 


2.29 


Slope = ky 
k H (nBu 3 SnH) = 


0.408 
//C ad d= 16.0 
2.1 (±0.3)x 


±0.1 
10 6 M 1 s" 1 


6.14 


100 


Pentafluoroethvl Radical (118) 











Ratios of 120 / 122 were determined the sum of integrals of the CF 3 - and -CF 2 H 
resonances of 120 (at <j> -86.6 and -138.6, respectively) versus that of the CF 3 - and -CF 2 - 
resonances of 122 (at <j> -86.1 and -119.2, respectively). 



Table 5-6. Data for Competitive Determination of k H (nBu 3 SnH) / /c add (Styrene) for 
Pentafluoroethyl Radical (118). 



0.082 


2.20 


0.434 


1.15 


89 


0.082 


2.08 


0.485 


1.31 


85 


0.082 


1.96 


0.542 


1.46 


85 


0.082 


1.84 


0.608 


1.62 


82 


0.082 


1.71 


0.682 


1.79 


83 






133 

Table 5-6- continued 

f 119 1 f C fi H s CH=CH ? 1 r nBihSnH 1 / rC fi Hs CH=CH, 1 f 1201 1\ 1221 % Yield 
0.082 1.59 0.768 2.05 83 

Slope = k H I k aM = 2.62 ± 0.06 
k H (nBu 3 SnH) = 3.2 (± 0.3) x 10 8 M" 1 s" 1 

1,1-Difluorohex-5-en-1-yl Radical (128) 

Ratios of 136 to 138 were determined by integration of the -CF 2 H resonance of 
136 (at <|> -116.2) versus the sum of integrals for the diastereotopic -CF 2 - resonances of 
138 (at <|> -100.3 and -107.8) in the 19 F NMR. 



Table 5-7. Data for Competitive Determination of k H (nBu 3 SnH) / k C5 for 1,1-Difluorohex- 
5-en-1-yl Radical (128). 

r 135 l LnBusSnHJ r 136 1/f 1381 % Yield 



0.054 


0.673 


1.53 


88 


0.054 


0.807 


1.91 


100 


0.054 


0.942 


2.28 


89 


0.054 


1.08 


2.57 


94 


0.054 


1.21 


2.93 


95 


0.054 


1.35 


3.29 


92 



Slope = k H I k C5 = 2.57 ± 0.05. 
/<c5 = 3.5(±0.6)x10 6 s- 1 

2,2-Difluorohex-5-en-1-vl Radical (140) 

Ratios of 145 to 147 and 145 to 149 were determined by integration of the 
-CH 2 CF 2 CH 3 resonance of 145 (at <j> -91.5) versus the sum of integrals for the 
diastereotopic -CF 2 - resonances of 147 (at <j> -88.9 and -90.4) and the -CF 2 - resonance 
of 149 (at + -95.8) in the 19 F NMR. 



134 



Table 5-8. Data for Competitive Determination of k H (nBu 3 SnH) / k C5 and k H (nBu 3 SnH) / 
/c C6 for2,2-Difluorohex-5-en-1-yl Radical (140). 



r 1441 


F nBu 3 SnH 1 
0.286 


[1451/ [1471 
2.64 


[14 


51/[1491 


% Yield 


0.053 




24.3 


98 


0.053 


0.343 


3.40 




30.9 


100 


0.053 


0.400 


4.03 




37.8 


100 


0.053 


0.457 


4.71 




48.2 


99 


0.053 


0.514 


5.44 




54.4 


98 


0.053 


0.572 


6.26 




61.1 


98 




For C 5 : 


Slope = k H 1 k C5 = 


= 12.5 


±0.2. 






/c C5 = 11 (±0.38)x 


10 6 s" 


i 





For C 6 : Slope = k H I k ce = 132 ± 5. 
/<C6 = 1.1 (±0.34)x10 5 s 1 



1,1,2,2-Tetrafluorohex-5-en-1-vl Radical (152) 

Ratios of 153 to 155 and 153 to 157 were determined by the sum of integrals of 
the -CF 2 - and -CF 2 H resonances of 153 (at § -1 16.7 and -135.9, respectively) versus the 
sum of integrals for the diastereotopic -CF 2 - resonances of 155 (at <)> -109.9, -120.8, 
-126.0, and -132.9) and the -CF 2 - resonance of 157 (at <|> -1 19.6, for k C6 ) in the 19 F NMR. 



Table 5-9. Data for Competitive Determination of k H ((TMS) 3 SiH) / k C5 and k H 
((TMS) 3 SiH) / k C6 for 1,1,2,2-Tetrafluorohex-5-en-1-yl Radical (152). 

[1011 f (TMS) 3 SiH 1 [ 153 1/f 1551 f 1531 /f 1571 % Yield 



0.052 


0.581 


0.988 


4.35 


96 


0.052 


0.686 


1.19 


5.28 


95 


0.052 


0.792 


1.44 


6.39 


94 


0.052 


0.897 


1.67 


7.36 


95 


0.052 


1.00 


1.84 


8.21 


98 


0.052 


1.11 


2.11 


9.37 


99 



135 



For C 5 : Slope = k H I k C5 = 2.11 ± 0.05. 

k C5 = 8.7 (± 0.41) x10 6 s 1 
For C 6 : Slope = k H I k C6 = 9.44 ±0.14. 

/c C6 = 1.9(±0.11)x10 6 s- 1 






APPENDIX A 
SELECTED 19 F NMR SPECTRA 

The 19 F NMR spectra of radical precursors, a representative sample from each 
kinetic run, and isolated products of addition, hydrogen abstraction, and cyclization are 
graphically illustrated in this appendix. Full spectral characterization data are presented 
numerically in their respective areas of Chapter 5. 



136 



137 




o 

CO 

<v 

c 

(0 
X 



o 

S 



6 

E 

2 
m 



< 



3 



138 










h* 



I 



c 
o 

1 

c 



•s 

c 
o 
Q. 
re 

1 

rsi 

< 

.g> 

u_ 






139 



( i 



S 



\ 






-t 



f! 



■ 
-S 






C 
CO 

s 

2 
o 

3 



CO 

< 

D 

ii. 



140 




-? 



. 



CO 
00 



CO 

I 

c 

<D 

-C 
Cl 
i 

i 



Q 
i 

CO 
CO 

i 
< 

13 



141 




CM 

C 
CO 
X 

Q) 

.c 

2 
o 

3 



csj 
c\f 

6 

■o 
o 



in 
i 

< 

C 



142 




oo 

1 

TO 

cr 



X 

<D 

jr 

2 
o 

3 

•E 
Q 

CN 

C\T 

i_ 

a 

c 
o 

TO 

C 



! 



*■ 



^ 



i 



: 



c 
o 

Q. 

TO 

+-^ 
TO 

Q 



CD 
i 

< 



OJ 



143 










<D 

C 

co 
x 

0) 

-C 
o 

o 

3 

S 

O 

i 

CM 
CM" 

< 

2> 









144 



I 

s 



-I 

K 



o 
I 










00 

a> 

c 
ra 
o 
o 

>» 
c 
a> 

Q. 

T— 
I 

2 
o 

b 
i 



oo 

I 

< 

CD 



145 




h 




CM 



C 
CO 
X 



2 
CM 

cm" 



r- >? 



o 

£ 

2 

m 



i 
< 

2 

Ll 



146 



^-i 



K'flMC \ 

SC'SrSZC 1 \ 

t9-5CKC- 

wactac— c 
96 »aszc- — r 

CSC9i2C 

>• 5Si3C- — j 

:s 5Gi2C-— 
C9 »Ci2C- — 
•C LZL7Z- — 
W'3Ii3C- 



:S 



r« 



_J 



J 



X 



J L, 



:a 



L« 




8 

ro 
K 



■ 



O 



CM 



C 

g 

c 



J 



£ 



c 
o 

ra 
Q 



< 

Ll 



147 



| 

■ a. 



Y ' 



00 

o 



: T 



c 
co 
x 

2 



o 

>•— 

I 

CM 
CM 



< 

Ll 









148 



! 0. 










t> 







s 






c 










o 


o 




1 


o 






>< 


[ 


c 


L 


£ 


L 

L ° 


si 


a. 
i 




T— 




o 




c 


L 


o 




D 




»•— 




03 




C 




•4-* 









1- 




■<3- 


-o 


■<r 




CO 






CO 



CN 



< 
O) 



149 



}2 



r? } 



•.a 



o 





c 

s 

O 

i 



3 

cm" 



CO 



< 



150 



-I 

a 







M 



C 
03 

i 

X 

0) 



• 



o 

_g 

t: 
aj 
a. 
■ 

CN 

6 

T3 
O 



< 



151 



i <*• 



*•: 



Zj 



IS ^ 



-? } 



to 
o 

TO 

a: 



di 



<a 



o 

a) 



C 

o 

'-•-^ 
TO 

C 

i 
S 



jf 






c 
o 
0. 

TO 

Q 
in 






< 



152 



-I 



=< ' } 






ID 



CO 











T 


— 






0) 

c 






5 






-C 






<u 




-co 

1 


>< 

X 












o 






1— 




r: 


o 




3 




o 
-co 

1 


9 






Q. 


- 






(d 




T— 




< 




0) 




I— 




D 




O) 








u_ 



153 



I 

s 



J 9 



-3 ^ - 






<D 

C 

c 


a 



■ 



2 
o 

t 


a. 



< 



154 



L a 



1 : 



o> 



L 

: 
L 



a) 

c 
ra 

.c 

I 

i— 

o 

c 

I 

o 



00 

I 

< 



155 



\l 



CO 

o 

CO 

I 

o 

i 

c 

CD 

CL 



c 
g 
**- 

CO 

c 
E 

0) 



^? 






c 
o 

CL 
CO 
CO 

Q 
oi 






< 



156 



I 



06»t36t-__ 



—* 



-? 



1-2 






o 





c 

03 

o 

1— 

o 

!= 

TO 

c 

! 



o 

< 



157 



| 



$ 






i 

I 

o 

C 


a. 

■ 

CM 

eg 



CM 

< 

ii. 



158 



-» 



-» 










-s 



in 




C 




X 





o 

3 



CD 
CO 

6 

E 
o 

I— 

m 

I 
CD 

CN 

CN 

< 



159 




3 




_: 



^a 



Li 



oo 



s 



i 
x 

<L> 
-C 
O 



Q 



g 

I 

c 

E 

a) 
Q 



j? 



c 
o 
0- 
ro 

■•-• 

(TJ 

Q 



CO 

CM 

< 



160 




Li 






c 




X 

<D 

.c 

2 
o 

i 

Q 

CD 

to" 



"J- 

CNJ 

I 

< 

D 

Ll 



161 




ti «>$? — ^ 



oo 
to 



0) 

c 

i 

c 

i 

o 
o 

.c 

0) 

E 

I 
CNJ 

2 
o 

3 



ID 
CM 

i 

< 



162 



| 

s 




5 



CD 

C 
<D 



X 



O 

3 



i r 



in 
6 

E 

2 
m 

CO 

CO 
CN 

i 

< 

3 

Ll 






163 




T3 
TO 



c 

i 
in 



o 

3 



CN 

cm" 



c 
o 

TO 

c 



2 
a 



£ 



c 
o 
Q. 

TO 

-*^ 
TO 

Q 



CM 

i 
< 






164 




I 



in 



(D 

C 



<D 

o 

t— 

o 

3 

b 
in 

in" 



oo 

CM 

< 

u 
O) 

Ll 



165 



| 

- a. 



►0»6SS2-_ 
Jf96SSZ-- 

10 1LSSZ- 




9t ooos: 
9S986»Z- 

I6Ci6»Z 



0) 

c 
a> 

c 

| 

o 

o 

>» 

.c 
1 

E 

■ 

CO 



o 

5= 

Q 



i 
< 

0) 

u. 
D 

Ll 






166 







a> 

c 

03 
X 

0) 



o 
o 

2 
o 



o 

< 

2 

3 



167 



. }'■ 




.( 



} = 



-? 




\-n 



0) 

c 



X 

<D 

JZ 

o 
o 

>•- 

ro 

i— 
*-» 

<D 

♦^ 
I 

CD 

to" 
in 

■ 
o 

E 

o 

i— 

GO 

I 

CO 



CO 

< 

3 



168 



: 



CM 



o 

CD 

a: 



I 



C 

CD 

i 

m 
■ 
x 

0) 





o 

E 

"53 



CN 

C\f 



C 

o 

c 



Q) 

Q 



J? 



c 
o 
0. 
ro 
fo 
Q 

csi 

CO 

I 

< 



L. 

3 



169 



I 



■ 






10 



a; 

c 

9 



S 









o 

3 

1 
« 

I- 

i 

03 

CO 

in 
uo 

CO 
CO 

< 



170 






3 



i 

*■ a. 



1 k: 



u> 



0> 

c 

TO 

c 

I 

o 
o 

I 

.c 

I 

E 

■ 

CO 



3 



o 

1 



CM 

(N 



CO 
I 

< 

3 



171 



U) 



c 

CO 
X 

0) 

-C 

o 
o 

>. 
o 
o 



1 

a3 



cm" 



m 

CO 

< 

s 



APPENDIX B 

B3LYP/6-31G(d) TOTAL AND ZERO-POINT ENERGIES 
FOR DATA IN TABLES 3-3 AND 3-4 

The total electronic and unsealed zero-point energies for hydrofluorocarbon 
radicals and closed shell species used in the determination of the bond dissociation 
energies reported in Tables 3-3 and 3-4 are provided in this appendix. In instances 
where more than one conformer is provided, that of lowest energy was used in the C-C 
and C-H BDE computation. 

Table B-1. Total and (Unsealed) Zero-Point Energies for Fluorinated Alkanes. 
Species E(B3LYP/6-31G(d)). a.u. ZPE, a.u. 



CH3CH3 


-79.8304 167 


0.075231 


CH3CF3 


-377.5549 235 


0.052872 


CH3CF2H 


-278.3015 940 


0.061076 


CF 3 CF 2 H 


-576.0077 790 


0.037905 


CH3CH2CH3 


-119.1442 464 


0.104110 


CH3CF2CH3 


-317.6263 447 


0.088479 


C/r3Crl2Cn3 


-416.8699 039 


0.081649 


Gr3CF2Cri3 


-615.3356 199 


0.065266 


CH3CH2CF2H 


-317.6163 135 


0.089851 


CH3CH2CF2H 


-317.6162 650 


0.090006 


CH3CF2CF2H b 


-516.0884 655 


0.073687 


CH3CF2CF2H a 


-516.0853 571 


0.073705 


Un3UH2CH2Cn3 


-158.4580 400 


0.132860 



172 



173 

Table B-1-- continued 

Species E(B3LYP/6-31G(d)). a.u. ZPE, a.u. 

CH 3 CH 2 CH 2 ClV -158.4567 065 0.132957 

CH3CH2CF2CIV -356.9408 361 0.117249 

CH 3 CH 2 CF 2 ClV -356.9401 189 0.117349 

CH 3 CF 2 CF 2 CH 3 C -555.4151664 0.101158 

CH 3 CF 2 CF 2 CH 3 d -555.4110 960 0.101141 

CF 3 CH 2 CH 2 CH 3 C -456.1836 478 0.110224 

CF 3 CH 2 CH 2 CH 3 " -456. 1 826 866 0. 1 1 0445 

3 Methyl and Hydrogen Gauche. b Methyl and Hydrogen Anti. c Methyl (or Methyl and 
Trifluoromethyl) Groups Anti. d Methyl (or Methyl and Trifluoromethyl) Groups Gauche. 

Table B-2. Total and (Unsealed) Zero-Point Energies for Fluorinated Radicals. 

Species E(B3LYP/6-31G(d)). a.u. ZPE, a.u. 



H 


-0.5002 728 





CH 3 


-39.8382 909 


0.029849 


CF 3 


-337.5509 879 


0.012153 


CH 3 CH 2 


-79.1578 663 


0.059647 


CH 3 CF 2 


-277.6352 216 


0.047481 


Cr 3 CH 2 


-376.8760 847 


0.037830 


CF 3 CF 2 


-575.3382 007 


0.024522 


CH 3 CH 2 CH 2 


-118.4713 699 


0.088731 


CH 3 CH 2 CH 2 


-118.4711 179 


0.088945 


CH 3 CH 2 CF 2 


-316.9496 940 


0.076574 


CH 3 CH 2 CF 2 


-316.9496 712 


0.076368 


CH 3 CF 2 CH 2 


-316.9493 447 


0.073435 


CH 3 CF 2 CH 2 


-316.9489 182 


0.073225 



174 



Table B-2-- continued 

Species 

CH3CF2CF2 
CH3CF2CF2 

CF3CH2CH2 

CF3CH2CH2 



E(B3LYP/6-31G(d)), a.u. ZPE. a.u. 

-515.4178 170 0.060271 

-515.4173 201 0.060276 

-416.1952 103 0.066154 

-416.1951 213 0.066242 



3 Radical p Orbital Aligned with p C-H Bond. B p C-C Bond Alignment. ° p C-F bond 
alignment. 









REFERENCES 

1. Gomberg, M. J. Am. Chem. Soc. 1900, 22, 757. 

2. Waters, W. A. The Chemistry of Free Radicals; Clarendon Press: Oxford, 1946. 

3. Walling, C. Free Radicals in Solution; Wiley: New York, 1957. 

4. Free Radicals; Kochi, J. K., Ed.; Wiley: New York, 1973. 

5. Kharasch, M. S.; Jensen, E. V.; Urry, W. H. Science 1945, 102, 128. 

6. Kharasch, M. S.; Jensen, E. V.; Urry, W. H. J. Am. Chem. Soc. 1946, 68, 1 100. 

7. Kharasch, M. S.; Jensen, E. V.; Urry, W. H. J. Am. Chem. Soc. 1947, 69, 1105. 

8. Beckwith, A. L. J. Chem. Soc. Rev. 1993, 143. 

9. Hart, D.J. Science 1984, 223, 883. 

10. Giese, B. Radicals in Organic Synthesis: Formation of Carbon-Carbon Bonds; 
Pergamon Press: Oxford, 1986. 

1 1 . Curran, D. P. Synthesis 1 988, 41 7. 

12. Curran, D. P. Synthesis 1988, 489. 

13. Jasperse, C. P.; Curran, D. P.; Feviz, T. L. Chem. Rev. 1991, 91, 1237. 

14. Dowd, P.; Zhang, W. Chem. Rev. 1993, 93, 2091. 

15. Stork, G. In Selectivity- A Goal for Synthetic Efficiency; Bartmann, W., Trost, B. 
M., Eds.; Verlag Chemie: Basel, 1984. 

16. Newcomb, M. Tetrahedron 1993, 49, 1151. 

17. Kuivila, H. G. Accts. Chem. Res. 1968, 1, 299. 

18. Neumann, W. P. Synthesis 1987, 665. 

19. Tedder, J. M.; Walton, J. C. Accts. Chem. Res. 1976, 9, 183. 

20. Benson, S. W. Thermochemical Kinetics; Wiley: New York, 1973. 

21. Fischer, H. In Substituent Effects in Radical Chemistry; Viehe, H. G., Janousek, 
Z., Merenyi, R., Eds.; Reidel: Dordrecht; 1986; NATO ASI Series C, 189, 123. 



175 



176 

22. Beckwith, A. L. J.; Pigou, P. E. Aust. J. Chem. 1986, 39, 77. 

23. Chatgilialoglu, C. Accts. Chem. Res. 1992, 25, 188. 

24. Fukui, K. Fortsch. Chem. Forsch. 1970, 15, 1. 

25. Fleming, I. Frontier Orbitals and Organic Chemical Reactions; Wiley: London, 
1976 

26. Burkey, T. J.; Majewski, M.; D. J. Am. Chem. Soc. 1986, 108, 2218. 

27. Chatgilialoglu, C; Ingold, K. U.; Scaiano, J. C. J. Am. Chem. Soc. 1981, 103, 
7739. 

28. Chatgilialoglu, C; Dickhaut, J.; Giese, B. J. Org. Chem. 1991, 56, 6399. 

29. Kanabus-Kaminska, J. M.; Hawari, J. A.; Griller, D.; Chatgililoglu, C. J. Am. 
Chem. Soc. 1987, 109, 5267. 

30. Lusztyk, J.; Maillard, B.; Lindsay, D. A.; Ingold, K. U. J. Am. Chem. Soc. 1983, 
105, 3578. 

31. Clark, K. B.; Griller, D. Organometallics 1991, 10, 746. 

32. Chatgilialoglu, C; Guerrini, A.; Lucarini, M. J. Org. Chem. 1992, 57, 3405. 

33. Franz, J. A.; Bushaw, B. A.; Alnajjar, M. S. J. Am. Chem. Soc. 1989, 111, 268. 

34. Newcomb, M.; Manek, M. B. J. Am. Chem. Soc. 1990, 112, 9662. 

35. Griller, D.; Kanabus-Kaminska, J. M.; Maccoll, A. J. Mol. Struct. 1988, 163, 125. 

36. Chatgilialoglu, C; Scaiano, J. C; Ingold, K. U. Organometallics 1982, 1, 466. 

37. Chatgilialoglu, C; Ingold, K. U.; Lusztyk, J.; Nazran, A. S.; Scaiano, J. C. 
Organometallics 1983, 2, 1332. 

38. Momillen, D. F.; Golden, D. M. Ann. Rev. Phys. Chem. 1982, 33, 493. 

39. Hartung, J.; Gallou, F. J. Org. Chem. 1995, 60, 6706. 

40. Laidler, K. Chemical Kinetics, Harper and Row: New York, 1987. 

41. Wong, M. W.; Radom, L. J. Phys. Chem. 1995, 99, 8582. 

42. Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L; Levin, R. D.; Mallard 
W. G. J. Phys. Chem. Ref. Data 1988, 17, Suppl. 1. 

43. Hammond, G. S. J. Am. Chem. Soc. 1955, 77, 334. 

44. Abell, P. I. In Comprehensive Chemical Kinetics; Bamford, C.H., Tipper, C. F. H., 
Eds.; Elsevier: Amsterdam, 1976, 18, Ch. 3, 111. 



177 



45. Hoyland, J. R. Helv. Chim. Acta 1971, 22, 229. 

46. Houk, K. N.; Paddon-Row, M. N.; Spellmeyer, D. C; Rondan, N. G.; Nagase, S. 
J. Org. Chem. 1986, 51, 2874. 

47. Zipse, H.; He, J.; Houk, K. N.; Giese, B. J. Am. Chem. Soc. 1991, 113, 4324. 

48. Wong, M. W.; Pross, A.; Radom, L J. Am. Chem. Soc. 1994, 116, 6284. 

49. Wong, M. W.; Pross, A.; Radom, L. J. Am. Chem. Soc. 1994, 116, 11938. 

50. Houk, K. N. Accts. Chem. Res. 1975, 8, 361. 

51. Heberger, K.; Walbiner, M.; Fischer, H. Angew. Chem., Int. Ed. Engl. 1992, 31, 
635. 

52. Tedder, J. M.; Walton, J. C. Tetrahedron 1980, 36, 701 . 

53. Riemenschneider, K.; Bartlels, H. M.; Dornow, R.; Drechsel-Grau, E.; Eichel, W.; 
Luthe, H.; Matter, Y. M.; Michaelis, W.; Boldt, P. J. Org. Chem. 1987, 52, 205. 

54. Avila, D. V.; Ingold, K. U.; Lusztyk, J.; Dolbier, W. R., Jr.; Pan, H.-Q.; Muir, M. J. 
Am. Chem. Soc. 1994, 116, 99. 

55. Giese, B.; He, J.; Mehl, W. Chem. Ber. 1988, 121, 2063. 

56. Santi, R.; Bergamini, F.; Citterio, A.; Sebastiano, R.; Nicolini, M. J. Org. Chem. 
1992, 57, 4250. 

57. Beranek, I.; Fischer. In Free Radicals in Synthesis and Biology; Minisci, F., Ed.; 
Kluwer: Dordrecht, 1989, NATO ASI Series C, 260, 303. 

58. Lamb, R. C; Ayers, P. W.; Toney, M. K. J. Am. Chem. Soc. 1963, 85, 3483. 

59. Baldwin, J. E. J. Chem. Soc, Chem. Commun. 1976, 734. 

60. Beckwith, A. L. J.; Moad, G. J. Chem. Soc, Chem. Commun. 1974, 472. 

61. Beckwith, A. L. J.; Ingold, K. U. In Rearrangements in Ground and Excited 
States; de Mayo, P., Ed.; Academic Press: New York, 1980, 1, Ch. 4, 161. 

62. Bischof, P. Helv. Chim. Acta 1980, 63, 1434. 

63. Beckwith, A. L. J.; Schiesser, C. H. Tet. Lett. 1985, 26, 373. 

64. Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron 1985, 41, 3925. 

65. Canadell, E. J. Chem. Soc, Perkin Trans I1 1985, 1331. 

66. Spellmeyer, D. C; Houk, K. N. J. Org. Chem. 1987, 52, 959. 

67. Delia, E. W.; Knill, A. M. Aust. J. Chem. 1995, 48, 2047. 



178 



68. Hartung, J.; Stowasser, R.; Vitt, D.; Bringmann, G. Angew. Chem. Int. Ed. Engl. 
1996, 35, 2820. 

69. Capon, B.; Rees, C. W. Ann. Rep. Chem. Soc. 1964, 61, 221. 

70. Beckwith, A. L. J.; Blair, I. A.; Phillipou, G. Tet. Lett. 1974, 2251. 

71. Beckwith, A. L. J.; Easton, C. J.; Lawrence, T.; Serelis, A. K. Aust. J. Chem. 
1983, 36, 545. 

72. Allinger, N. L; Zalkow, J. J. Org. Chem. 1960, 25, 701. 

73. Lusztyk, J.; Maillard, B.; Deycard, S.; Lindsay, D. A.; Ingold, K. U. J. Org. Chem. 
1987, 52, 3509. 

74. Park, S.-U.; Chung, S.-K.; Newcomb, M. J. Am. Chem. Soc. 1986, 108, 240. 

75. Johnson, C. C; Horner, J. H.; Tranche, C; Newcomb, M. J. Am. Chem. Soc. 
1995, 117, 1684, 

76. Newcomb, M.; Horner, J. H.; Filipkowski, M. A.; Ha, C; Park, S.-U. J. Am. Chem. 
Soc. 1995, 117, 3674. 

77. Horner, J. H.; Martinez, F. N.; Musa, O. M.; Newcomb, M.; Shahin, H. J. Am. 
Chem. Soc. 1995, 1 1 7, 11124. 

78. Khalil, S. M.; Jarjis, H. M. Z Naturforsch., Sec. A. 1991, 46, 898. 

79. Korth, H. G.; Lommes, P.; Sustmann, R. J. Am. Chem. Soc. 1984, 106, 663. 

80. Pasto, D. J. J. Am. Chem. Soc. 1988, 110, 8164. 

81. Walling, C; Cioffari, A. J. Am. Chem. Soc. 1972, 94, 6064. 

82. Julia, M.; Maumy, M. Bull. Soc. Chim. Fr. 1966, 434. 

83. Kazanis, S.; Azarani, A.; Johnston, L. J. Phys. Chem. 1991, 95, 4430. 

84. Barton, D. H. R.; Crich, D.; Motherwell, W. B. Tetrahedron 1985, 41, 3901. 

85. Avila, D. V.; Ingold, K. U.; Lusztyk, J.; Dolbier, W. R., Jr.; Pan, H.-Q. J. Am. 
Chem. Soc. 1993, 115, 1577. 

86. Rong, X. X.; Pan, H.-Q.; Dolbier, W. R., Jr. J. Am. Chem. Soc. 1994, 116, 4521. 

87. Dolbier, W. R., Jr.; Rong, X. X. J. Fluorine Chem. 1995, 72, 235. 

88. Avila, D. V.; Ingold, K. U.; Lusztyk, J.; Dolbier, W. R., Jr.; Pan, H.-Q. J. Org. 
Chem. 1996, 61, 2027. 

89. Dolbier, W. R., Jr.; Rong, X. X.; Smart, B. E.; Yang, Z.-U. J. Org. Chem. 1996, 
61, 4824. 



179 



90. Dolbier, W. R., Jr. Chem. Rev. 1996, 96, 1557. 

91. Bamberger, M. D.; Dolbier, W. R., Jr.; Lusztyk, J.; Ingold, K. U. Tetrahedron 1997, 
53, 9857. 

92. The Chemist's Companion; Gordon, A. J.; Ford, R. A.; Eds., Wiley: New York, 
1972 

93. Bondi, A. J. Phys. Chem. 1964, 68, 441 

94. Smart, B. E. In Organofluorine Chemistry: Principles and Commercial 
Applications; Banks, R. E., Smart, B. E., Tatlow, J. C, Eds.; Plenum: New York, 
1994, Ch. 3, 57. 

95. Smart, B. E. In Molecular Structure and Energetics; Liebman, J. F., Greenberg, 
A., Eds.; VCH: Deerfield Beach, FL, 1986, 3, Ch. 4, 141. 

96. Smart, B. E. In The Chemistry of Functional Groups, Supplement D; Part 2; 
Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1983, 603. 

97. Smart, B. E. In Chemistry of Organic Fluorine Compounds II: A Critical Review; 
Hudlicky, M., Pavlath, A. E., Eds.; ACS: Washington, 1995, Ch. 6, 979. 

98. Pauling, L. The Nature of the Chemical Bond; Cornell University Press: Ithaca, 
1960 

99. Radom, L; Hehre, W. J.; Pople, J. A. J. Am. Chem. Soc. 1971, 93, 289. 

100. Radom, L; Stiles, P. J. Tetrahedron Lett. 1975, 789. 

101. Baird, N. C. Can. J. Chem. 1983, 61, 1567. 

102. Typke, V.; Dakkouri, M;. Oberhammer, H. J. Mol. Struct. 1978, 44, 85. 

103. Harmony, M. D.; Laurie, V. W.; Kuczkowski, R. L; Schwendeman, R. H.; 
Ramsay, D. A.; Lovas, F. J.; Lafferty, W. J.; Maki, A. G. J. Phys. Chem. Ref. Data 
1979, 8, 619. 

104. Wiberg, K. B.; Rablen, P. R. J. Am. Chem. Soc. 1993, 115, 614. 

105. Bock, C. W.; George, P.; Mains, G. J.; Trachtman, M. J. Chem. Soc. Perkin 
Transit 1979, 814. 

106. Facelli, J. C; Contreras, R. H. Z. Naturforsch. 1980, 35A, 1350. 

107. Bak, B.; Kierkegaard, C; Pappas, J.; Skaarup, S. Acta Chem. Scand. 1973, 27 
363. 

108. Bernett, W. A. J. Org. Chem. 1969, 34, 1772. 

109. Kollman, P. J. J. Am. Chem. Soc. 1974, 96, 4363. 






180 



110. Epiotis, N. D.; Cherry, W. R.; Shaik, S.; Yates, R. L; Bernardi, F. Top. Curr. 
Chem. 1977, 70, 1. 

111. Epiotis, N. D.; Larson, J. R.; Eaton, H. L. Unified Valence Bond Theory of 
Electronic Structure, Springer- Verlag: New York, 1982, Lecture Notes in 
Chemistry, 29. 

112. Brundle, C. R.; Robin, M. B.; Kuebler, N. A.; Basch, H. J. Am. Chem. Soc. 1972, 
94, 1451. 

113. Chiu, N. S.; Burrow, P. D.; Jordan, K. D. Chem. Phys. Lett. 1979, 68, 121. 

1 14. Dolbier, W. R.; Jr.; Medinger, K. S. Tetrahedron 1982, 38, 2415. 

115. Chambers, R. D. Fluorine in Organic Chemistry, Wiley: New York, 1973. 

116. Bernardi, F.; Mangini, A.; Epiotis, N. D.; Larson, J. R.; Shaik, S. J. Am. Chem. 
Soc. 1977, 99, 7465. 

117. Blint, R. J.; McMahon, T. B.; Beauchamp, J. L J. Am. Chem. Soc. 1974, 96, 
1269. 

118. Williams, A. D.; Le Breton, P. R.; Beauchamp, J. L. J. Am. Chem. Soc. 1976, 98, 
2705. 

119. Olah, G. A.; Mo. Y. K. Adv. Fluorine Chem. 1973, 7, 69. 

120. Prochaska, F. T.; Andrews, L. J. Am. Chem. Soc. 1978, 100, 2102. 

121. Olah, G. A.; Heiliger, L; Prakash, G. K. S. J. Am. Chem. Soc. 1989, 111, 8020. 

122. Dixon, D. A.; Fukunaga, T.; Smart, B. E. J. Am. Chem. Soc. 1986, 108, 4027. 

123. Dixon, D. A.; Fukunaga, T.; Smart, B. E. J. Phys. Org. Chem. 1988, 1, 153. 

124. Marynick, D. S. J. Mol. Struct. 1982, 87, 161. 

125. Reutov, O. A.; Beletsyaka, I. P.; Butin, K. P. CH-Acids; Pergamon: Oxford, 1978. 

126. Streiweiser, A., Jr.; Mares, F. J. Am. Chem. Soc. 1968, 90, 2444. 

127. Farnham, W. B.; Smart, B. E.; Middleton, W. J.; Calabrese, J. O; Dixon, D. A. J. 
Am. Chem. Soc. 1985, 707, 4565. 

128. Farnham, W. B.; Dixon, D. A.; Calabrese, J. C. J. Am. Chem. Soc. 1988, 110, 
2607. 

129. Cosmo, R.; Stemhell, S. Aust. J. Chem. 1987, 40, 35. 

130. Sherrod, S. A.; daCosta, R. L; Barnes, R. A., Boekelheide, V. J. Am. Chem. Soc. 
1974, 96, 1565. 

131. Dolbier, W. R., Jr.; Palmer, K. W. J. Am. Chem. Soc. 1993, 115, 9349. 



181 



132. Palmer, K. W. Ph.D. Dissertation, University of Florida, 1993. 

133. Hirsch, J. Top. Sterochem. 1967, 7, 199. 

134. Hansch, C; Leo, A. Substituent Constants for Correlation Analysis in Chemistry 
and Biology; Wiley: New York, 1979. 

135. Dawson, W. H.; Hunter, D. H.; Willis, C. J. J. Chem. Soc, Chem. Commun. 
1980, 874. 

136. Scherer, K. V., Jr.; Ono, T.; Yamanouchi, R.; Fernandez, R.; Henderson, P.; 
Goldwhite, H. J. Am. Chem. Soc. 1985, 707, 718. 

137. Haszeldine, R. N. J. Chem. Soc. 1949, 2856. 

138. Tarrant, P.; Lovelace, A. M. J. Am. Chem. Soc. 1954, 76, 3466. 

139. Tarrant, P.; Lovelace, A. M. J. Am. Chem. Soc. 1955, 77, 2783. 

140. Stefani, A. P.; Herk, L; Szwarc, M. J. Am. Chem. Soc. 1961, 83, 4732. 

141. Tedder, J. M.; Walton, J. C; Winton, K. D. R. J. Chem. Soc, Faraday Trans. I 
1972, 68, 1. 

142. Low, H. C; Tedder, J. M.; Walton, J. C. J. Chem. Soc, Faraday Trans. I 1976, 
72, 1707. 

143. Krusic, P. J.; Bingham, R. C. J. Am. Chem. Soc. 1976, 98, 230. 

144. Bernardi, F.; Cherry, W.; Shaik, S.; Epiotis, N. D. J. Am. Chem. Soc. 1978, 700, 
1352. 

145. Benzel, M. A.; Maurice, A. M.; Belford, R. L; Dykstra, C. E. J. Am. Chem. Soc. 
1983, 705, 3802. 

146. Paddon-Row, M. N.; Wong, S. S. J. Mol. Struct. 1987, 750, 109. 

147. Pasto, D. J.; Krasnansky, R.; Zercher, C. J. Org. Chem. 1987, 52, 3062. 

148. Wang, S. Y.; Borden, W. T. J. Am. Chem. Soc. 1989, 777, 7282. 

149. Paddon-Row, M. N.; Thomson, C; Ball, J. R. J. Mol. Struct. 1987, 750, 93. 

150. Paddon-Row, M. N.; Wong, S. S. J. Mol. Struct. 1987, 750, 109. 

151. Chen, Y.; Rauk, A.; Tschuikow-Roux, E. J. Chem. Phys. 1990, 93, 1187. 

152. Chen, Y.; Rauk, A.; Tschuikow-Roux, E. J. Chem. Phys. 1990, 93, 6620. 

153. Paddon-Row, M. N.; Houk, K. N. J. Phys. Chem. 1985, 89, 3771. 

154. Bordwell, F. G.; Zhang, X. M. Ace Chem. Res. 1993, 26, 510. 



182 



155. Seakins, P. W.; Pilling, M. J.; Niiranen, J. T.; Gutman, D.; Krasnoperov, L. N. J. 
Phys. Chem. 1992, 96, 9847. 

156. Pickard, J. M.; Rodgers, A. S. J. Am. Chem. Soc. 1977, 99, 691. 

157. Martell, J. M.; Boyd, R. J.; Shi, Z. J. Phys. Chem. 1993, 97, 7208. 

158. Marshall, P.; Schwartz, M. J. Phys. Chem. A 1997, 101, 2906. 

159. Zachariah, M. R.; Westmoreland, P. R.; Burgess, D. R., Jr.; Tsang, W.; Melius, 
C. F. J. Phys. Chem. 1996, 100, 8737. 

160. Koopmans, T. A. Physica 1933, 7, 104. 

161. Rodriquez, C. F.; Sirois, S.; Hopkinson, A. C. J. Org. Chem. 1992, 57, 4869. 

162. Danovich, D.; Apeloig, Y.; Shaik, S. J. Chem. Soc, Perkin Trans. I1 1993, 321. 

163. Tedder, J. M. Tetrahedron 1982, 38, 313. 

164. Kim, S. S.; Kim, S. Y.; Ryou, S. S.; Lee, S. S.; Yoo, K. H. J. Org. Chem. 1993, 
58, 192. 

165. Brace, N. O. J. Am. Chem. Soc. 1964, 86, 523. 

166. Brace, N. O. J. Org. Chem. 1966, 31, 2879. 

167. Smart, B. E.; Feiring, A. E.; Krespan, C. G.; Yang, Z.-Y.; Hunh, M.-H.; Resnick, 
P. R.; Dolbier, W. R., Jr.; Rong, X. X. Macromol. Symp. 1995, 98, 753. 

168. Hung, M.-H.; Resnick, P. R.; Smart, B. E.; Buck, W. H. In Polymeric Materials 
Encyclopedia; Salamone, J. C, Ed.; CRC Press: Boca Raton, 1996, 4, 2466. 

169. Biomedical Aspects of Fluorine Chemistry; Filler, R., Kobayashi, Y., Eds.; 
Kodansha and Elsevier Biomedical: Tokyo, Amsterdam and New York, 1983. 

170. Rong, X. X. Ph.D. Dissertation, University of Florida, 1995. 

171. Gonzalez, J.; Foti, C. J.; Elsheimer, S. J. Org. Chem. 1991, 56, 4322. 

172. Burton, D. J.; Kehoe, L. J. J. Org. Chem. 1970, 35, 1339. 

173. Zav'yalov, S. I.; Ezhova, G. I.; Sitkareva, I. V.; Dorofeeva, O. V.; Zavozin, A. G ; 
Rumyantseva, E. E. Isv. Akad. Nauk. SSSR 1989, 10, 2392. 

174. Barton, D. H. R.; Lacher, B.; Zard, S. Z. Tetrahedron 1986, 42, 2325. 

175. Hauptschein, M.; Stokes, C. S.; Grosse, A. V. J. Am. Chem. Soc. 1952, 74, 848. 

176. Hauptschein, M.; Kinsman, R. L; Grosse, A. V. J. Am. Chem Soc 1952 74 
849. ' ' 



183 



177. Hudlicky, M. Chemistry of Organic Fluorine Compounds; Ellis Horwood: 
Chichester, 1976, 358. 

178. Pierce, O. R.; Meiners, A. F.; McBee, E. T. J. Am. Chem. Soc. 1953, 75, 2516. 

179. Miller, W. T., Jr.; Bergman, E.; Fainberg, A. H. J. Am. Chem. Soc. 1957, 79, 
4159. 

180. Alvernhe, G.; Laurent, A.; Haufe.G. Synthesis 1987, 562. 

181. Rondestvedt, C. S., Jr. J. Org. Chem. 1977, 42, 1985. 

182. Citterio, A.; Arnoldi, A.; Minisci, F. J. Org. Chem. 1979, 44, 2674. 

183. Johnston, L. J.; Scaiano, J. C; Ingold, K. U. J. Am. Chem. Soc. 1984, 706, 4877. 

184. Martell, J. M.; Boyd, R. J. J. Phys. Chem. 1992, 96, 6287. 

185. Becke, A. D. J. Chem. Phys. 1993, 98, 5648. 

186. Lee, C; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. 

187. CRC Handbook of Chemistry and Physics, 76 th Edition; Lide, D. R., Ed.; CRC 
Press: Boca Raton, 1995-1996, 9-63-9-67. 

188. MS 7 Structures Properties Database and Estimation Program 1991; U.S. 
Department of Commerce: Gaithersburg, MD, 1991. 

189. Wong, M. W.; Pross, A.; Radom, L J. Am. Chem. Soc. 1993, 115, 11050. 

190. Wong, M. W.; Pross, A.; Radom, L. Isr. J. Chem. 1993, 33, 415. 

191. Singh, U. C; Kollman, P. A. J. Comp. Chem. 1984, 5, 129. 

192. Besler, B. H.; Merz, K. M., Jr.; Kollman, P. A. J. Comp. Chem. 1990, 11, 431. 

193. Cort, A. D. Syn. Comm. 1990, 20, 757. 

194. Corey, E. J.; Su, W. Tet. Lett. 1984, 25, 51 19. 

195. Morikawa, T.; Kodama, Y.; Uchida, J.; Takano, M.; Washio, Y.; Taguchi, T. 
Tetrahedron 1992, 48, 8915. 

196. Nobes, R. H.; Pople, J. A.; Radom, L; Handy, N. C; Knowles, P. J. Chem. Phys. 
Lett. 1987, 738,481. 

197. Stanton, J. F. J. Chem. Phys. 1994, 707, 371. 

198. Chen, W.; Schlegel, H. B. J. Chem. Phys. 1994, 707, 5957. 






184 



199. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; 
Robb, M. A.; Cheeseman, J. R.; Keith T.; Petersson, G. A.; Montgomery, J. A.; 
Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. 
B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. 
Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; 
Gomperts, R.; Martin, R. L, Fox, D. J., Binkley, J. S.; Defrees, D. J.; Baker, J.; 
Stewart, J. P.; Head-Gordon, M.; Gonzalez, C; Pople, J. A. Gaussian 94, 
Revision C.3, Gaussian, Inc., Pittsburgh PA, 1995. 

200. Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502. 

201. Bauschlicher, C. W., Jr.; Partridge, H. J. Chem. Phys. 1995, 103, 1788. 

202. Hehre, W. J.; Radom, L; Schleyer, P. v. R.; Pople, J. A. Ab Initio Molecular 
Orbital Theory, Wiley: New York, 1986. 






BIOGRAPHICAL SKETCH 

Michael David Bartberger was born September 6, 1970, in Ft. Lauderdale, 
Florida, the first of two children, and spent his childhood years in the Margate / North 
Lauderdale area. 

Michael received a B.S. degree in chemistry from the University of Central 
Florida in May of 1992, during which time he was introduced to organic fluorine 
chemistry by University of Florida alumnus and former Dolbier group member, Professor 
Seth Elsheimer. 

In August of 1992, Michael began graduate study at the University of Florida, 
having practically joined the research group of Professor William R. Dolbier, Jr. even 
before moving to Gainesville. In addition to experimental physical organic studies, 
Michael developed an interest in theoretical methodology, having been exposed to the 
results of AM1 calculations attractively displayed on a Tektronix CAChe workstation. 
With a few days of instruction from Dr. Max Muir (a postdoc from another group working 
on some semiempirical calculations as a favor to the Dolbier group, leaving UF shortly 
thereafter for a position at MSI) Michael set out to hone his own computational chemistry 
skills. Spending many late nights at the workstation, he developed his proficiency with 
the MOPAC, ZINDO, Gaussian, ACES II, and GAMESS program systems, leading to a 
number of productive collaborations with other members of the department, both within 
and outside the Dolbier group. The highlight of his graduate career came in August of 
1996, as an invited participant in the Elucidation of Reaction Mechanisms by Ab Initio 
Methods Symposium held at the 212 th National Meeting of the American Chemical 
Society in Orlando, Florida. 



185 



186 

Michael is a recipient of the Department of Chemistry Excellence in Teaching 
Award, the Shell Foundation Fellowship, and has twice received the Outstanding 
Presentation Award administered by the Florida Section of the American Chemical 
Society, once as an undergraduate. 

At the time of this writing, Michael had accepted a position as a postdoctoral 
research associate in the Department of Chemistry and Biochemistry at the University of 
California, Los Angeles, under the direction of Professor K. N. Houk. 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 



(l/jJUJk 





William R. Dolbier, Jr., Chair 
Professor of Chemistry 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 



^~)u«A G- £?>z&£& 



Merle A. Battiste 
Professor of Chemistry 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 



Kirk S. Schanze 
Professor of Chemistry 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 



tfy<u£L?& 



Rodney J. Bai^lett 
Graduate Research Professor of 
Chemistry 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 




JgWh R. Sabin 
Professor of Physics 



This dissertation was submitted to the Graduate Faculty of the Department of 
Chemistry in the College of Liberal Arts and Sciences and to the Graduate School and 
was accepted as partial fulfillment of the requirements for the degree of Doctor of 
Philosophy. 

May, 1998 



Dean, Graduate School 






LD 

1780 
199Q 






UNIVERSITY OF FLORIDA 



3 1262 08555 1058